Logistics scheduling method and system for e-commerce trade

Abstract

The invention provides a logistics scheduling method and system for e-commerce trade, and relates to the technical field of computers, and the method comprises the steps: obtaining remote parcel logistics information, and obtaining at least one piece of hitchhiking reservation list information of which the corresponding travel destination is a remote region; performing preliminary screening on each piece of hitchhiking reservation list information according to the package attribute, the package expected arrival time, the package delivery address and the distribution station address so as to determine at least one piece of matched candidate hitchhiking reservation list information; determining target hitchhiking reservation order information from the candidate hitchhiking reservation order information based on a logistics scheduling matching model; and according to the distribution station address and the parcel delivery address, a driving path in the target hitchhiking reservation list information is updated, and a logistics scheduling instruction is sent to a corresponding vehicle owner client. Therefore, the parcels in the remote area are transported by adopting hitchhiking cooperative scheduling, so that the distribution time is greatly shortened, and the transportation cost can be greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a logistics scheduling method and system for e-commerce trade. Background Art

[0002] With the rapid development of e-commerce, the logistics and distribution system has gradually become one of the core supports for the efficient operation of e-commerce business. However, in the whole process of logistics and distribution, the "last mile" distribution link is particularly critical, which directly determines the quality of user experience and the success of logistics services.

[0003] At present, for the "last mile" delivery, delivery stations are set up within a certain area, and the packages are centrally delivered to the stations, and then the delivery personnel complete the door-to-door delivery. Alternatively, smart express lockers are deployed in areas with concentrated residents, and users go to the express lockers to pick up the packages. Because they can make full use of the scale effect and reduce the cost of single-piece delivery, these models are effective in high-density urban areas.

[0004] However, compared with cities, the population in remote areas is dispersed, the number of packages is limited, and it is difficult to form economies of scale. The delivery distance of a single package is long, and some terrains are complex (such as mountainous areas and hills), resulting in long delivery time and high transportation costs. Setting up distribution stations or dispatching full-time delivery personnel for remote areas alone is costly and difficult to maintain in the long term. In order to reduce costs, logistics companies usually adopt periodic distribution (such as centralized delivery once a week), which significantly prolongs the waiting time for users.

[0005] In response to the above problems, the industry has not yet proposed a better technical solution. Summary of the invention

[0006] The embodiment of the present invention provides a logistics scheduling method and system for e-commerce trade, which are used to at least solve the problems of long logistics delivery time and high transportation cost in remote areas in the prior art.

[0007] In a first aspect, an embodiment of the present invention provides a logistics scheduling method for e-commerce trade, comprising: obtaining remote package logistics information, the remote package logistics information including the package delivery address, package attributes, time urgency and expected package arrival time of the corresponding remote area; obtaining at least one ride-sharing reservation order information corresponding to the remote area as the travel destination, the ride-sharing reservation order information including basic vehicle information, driving route, departure time and logistics scheduling historical order records; performing preliminary screening of each of the ride-sharing reservation order information according to the package attributes, the expected package arrival time, the package delivery address and the distribution station address to determine at least one matching candidate ride-sharing reservation order information; inputting each of the candidate ride-sharing reservation order information and the remote package logistics information into a logistics scheduling matching model to determine a target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information; updating the driving route in the target ride-sharing reservation order information according to the distribution station address and the package delivery address, and sending a logistics scheduling instruction to the corresponding vehicle owner client.

[0008] In a second aspect, an embodiment of the present invention provides a logistics scheduling system for e-commerce trade, comprising: a first acquisition unit, used to acquire remote package logistics information, the remote package logistics information includes the package delivery address, package attributes, time urgency and expected package arrival time of the corresponding remote area; a second acquisition unit, used to acquire at least one ride-sharing reservation order information corresponding to the travel destination of the remote area, the ride-sharing reservation order information includes basic vehicle information, driving route, departure time and logistics scheduling historical order records; a reservation order initial screening unit, used to initially screen each of the ride-sharing reservation order information according to the package attributes, the expected package arrival time, the package delivery address and the distribution station address, so as to determine at least one matching candidate ride-sharing reservation order information; a scheduling matching unit, used to input each of the candidate ride-sharing reservation order information and the remote package logistics information into a logistics scheduling matching model, so as to determine the target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information; a scheduling execution unit, used to update the driving route in the target ride-sharing reservation order information according to the distribution station address and the package delivery address, and send a logistics scheduling instruction to the corresponding car owner client.

[0009] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the above method.

[0010] In a fourth aspect, an embodiment of the present invention provides a storage medium, in which one or more programs including execution instructions are stored. The execution instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to execute the steps of the above-mentioned method of the present invention.

[0011] In a fifth aspect, an embodiment of the present invention further provides a computer program product, wherein the computer program product comprises a computer program stored on a storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer executes the steps of the above method.

[0012] Compared with the prior art, the above technical solution has at least the following beneficial effects:

[0013] In the embodiments of the present invention, by combining ride-sharing scheduling with the traditional delivery model, the bottlenecks that the traditional model cannot adapt to, such as dispersed populations in remote areas and limited numbers of packages, are broken. By acquiring remote package information and ride-sharing reservation information in real time, accurate screening can be performed based on factors such as package attributes, expected arrival time, and delivery address, and idle ride-sharing vehicles can be matched with suitable packages to minimize delivery costs and improve delivery efficiency. Therefore, compared with the traditional method of individually dispatching full-time delivery personnel or periodic delivery, the use of ride-sharing collaborative scheduling not only greatly shortens the delivery time, but also greatly reduces transportation costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0015] Figure 1 A flowchart showing an example of a logistics scheduling method for e-commerce trade according to an embodiment of the present invention;

[0016] Figure 2 A flowchart showing an example of a logistics scheduling method for e-commerce trade according to an embodiment of the present invention;

[0017] Figure 3 A flowchart showing an example of performing preliminary screening of each ride-sharing reservation order information according to an embodiment of the present invention;

[0018] Figure 4 An operation flow chart showing an example of determining target rideshare reservation information from various candidate rideshare reservation information based on a logistics scheduling matching model according to an embodiment of the present invention;

[0019] Figure 5 A structural block diagram showing an example of a logistics scheduling system for e-commerce trade according to an embodiment of the present invention;

[0020] Figure 6 It is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one", "one" or "the" do not indicate a quantitative limitation, but indicate the existence of at least one. Words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0023] It should be noted that the terms "up", "down", "left", "right", "front", "back", etc. used in the present invention are only used to indicate relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly.

[0024] Figure 1 A flowchart showing an example of a logistics scheduling method for e-commerce trade according to an embodiment of the present invention is shown.

[0025] Regarding the execution subject of the method of the embodiment of the present invention, it can be any controller or processor with computing or processing capabilities to intelligently match the delivery demand of remote packages with the existing ride-sharing reservation information, and through intelligent scheduling, select the most suitable ride-sharing for delivery based on factors such as the time urgency of the package and the expected arrival time. In the current context of increasingly affordable and universal cars, by combining ride-sharing resources with the logistics and distribution needs of remote areas, the logistics and distribution costs and delivery time efficiency of remote areas are effectively reduced.

[0026] In some examples, it may be integrated and configured in the e-government system server through software, hardware, or a combination of software and hardware, and the type of the e-government system server may be diverse, such as a mobile phone, tablet computer, or desktop computer, etc.

[0027] like Figure 1 As shown, in step S110, remote package logistics information is obtained, and the remote package logistics information includes the package delivery address, package attributes, time urgency and expected arrival time of the package corresponding to the remote area.

[0028] In some embodiments, the logistics information about the remote package is obtained in real time by connecting with the API interface of the e-commerce platform or logistics management system. Specifically, the package delivery address is the geographical coordinates or address of the user or the receiving location; the package attributes include the volume, weight, whether it is fragile, and other characteristics of the package, which are helpful for considering the delivery method and cargo requirements in subsequent scheduling; the time urgency can be based on the delivery method selected by the user or the priority set by the merchant, identifying the delivery time requirements of the package, such as normal, expedited, etc.; the expected arrival time of the package is the time when the package is expected to arrive at the designated delivery station, which can be generated by the e-commerce platform or logistics service provider when confirming the delivery plan.

[0029] Preferably, the acquired parcel information is formatted and verified, and the acquired logistics information is stored in a database or memory cache to facilitate subsequent processing by the scheduling algorithm.

[0030] In step S120, at least one ride-sharing reservation order information corresponding to a trip destination in a remote area is obtained, and the ride-sharing reservation order information includes basic vehicle information, driving route, departure time and logistics scheduling historical order records.

[0031] In some embodiments, the reservation order information of the ride-sharing is obtained from the ride-sharing platform (or related vehicle management system), and the ride-sharing reservation order information related to the target remote area is queried and obtained according to the distribution demand of the target remote area. Specifically, the basic information of the vehicle may include the owner information, the vehicle type (size, load capacity); the driving route is the scheduled driving route of the ride-sharing driver, including the departure and destination, as well as the main sections passed along the way, and can be obtained through the map API or the itinerary information uploaded by the owner; the departure time is the time when the vehicle is scheduled to depart, which is used to judge the availability and time arrangement of the ride-sharing; the logistics scheduling history order record is the record of the vehicle's past collaborative completion of logistics scheduling, including the number of completed orders, service evaluation and other information, which is helpful to evaluate the scheduling reliability and efficiency of the ride-sharing.

[0032] In step S130, each ride-sharing reservation order information is preliminarily screened based on the package attributes, expected package arrival time, package delivery address and delivery station address to determine at least one matching candidate ride-sharing reservation order information.

[0033] In some embodiments, a rule-based matching algorithm is used to perform a preliminary screening of each ride-sharing reservation order to remove vehicles that clearly do not meet the requirements. For example, each package dimension is matched with the ride-sharing reservation order information, and each matching result is combined for preliminary screening, and the qualified ride-sharing reservation orders are retained as a candidate list.

[0034] For example, based on the size and weight of the package, select the rideshare that can carry the package. For example, if the package is large or heavy, select a vehicle with a strong load capacity or an empty vehicle. Select a rideshare whose departure time is later than or close to the expected arrival time of the package to ensure that the rideshare can dispatch the package. Match the package delivery address and distribution station address with each rideshare route, and filter out rideshares whose driving routes are seriously deviated from the package delivery address.

[0035] In step S140, each candidate ride-sharing reservation order information and remote package logistics information are input into a logistics scheduling matching model to determine the target ride-sharing reservation order information from each candidate ride-sharing reservation order information.

[0036] It should be noted that the logistics scheduling matching model can use various types of machine learning algorithms to optimize scheduling decisions, such as reinforcement learning, linear programming, integer programming, genetic algorithms, etc., to automatically screen the best rideshare, which is not limited here. Preferably, it is also possible to design a multi-objective optimization function, comprehensively considering factors such as cost, timeliness and path optimization, so that it not only considers the delivery cost (for example, the shortest driving distance or time), but also considers the delivery timeliness (for example, whether the package arrival time meets expectations) and rideshare scheduling reliability, etc., and balances between multiple goals to ensure that the selected rideshare can not only collaborate to complete logistics transportation, but also ensure the lowest transportation cost.

[0037] In step S150, the driving route in the target ride-sharing reservation order information is updated according to the distribution station address and the package delivery address, and a logistics dispatch instruction is sent to the corresponding vehicle owner client.

[0038] In some embodiments, according to the distribution station address of the target ride-sharing and the package delivery address, the map API (such as Baidu Map, Gaode Map, etc.) is used to calculate the optimal path through GIS and dynamic traffic data, so as to re-plan the driving route and avoid congested or impassable sections. Then, a logistics dispatch instruction is generated, including the vehicle's driving path, scheduled delivery time, package information, etc., and forwarded to the corresponding owner client through the owner information or the ride-sharing platform API. Preferably, during the delivery process, the driving status of the ride-sharing is monitored in real time, such as route deviation, traffic jam, etc., and the corresponding vehicle monitoring information is synchronized to the logistics tracking information.

[0039] Through the embodiment of the present invention, by obtaining the information of the ride-sharing reservation order and matching it with the package delivery demand, the maximum utilization of logistics resources (such as ride-sharing) is achieved. Since the basic information of the vehicle, the driving path, the departure time and other factors are accurately scheduled, the waste of vehicle resources or unnecessary empty driving is effectively avoided, which not only optimizes the logistics process, but also improves the utilization rate of social resources. Therefore, combined with the dynamic scheduling of the ride-sharing, the distribution plan can be flexibly adjusted according to the actual driving path of the vehicle. In the complex terrain of remote areas (such as mountainous areas, hills, etc.), the ride-sharing drivers are generally local residents. With their familiar geographical knowledge and driving flexibility, they can effectively improve the timeliness and success rate of distribution.

[0040] Figure 2 A flowchart showing an example of a logistics scheduling method for e-commerce trade according to an embodiment of the present invention is shown.

[0041] like Figure 2 As shown, in step S210, a logistics scheduling instruction is sent to the corresponding vehicle owner client.

[0042] For details on the implementation of step S210, please refer to the above combined Figure 1 The description of is not repeated here.

[0043] In step S220, it is detected whether the feedback for the logistics scheduling instruction is a confirmation feedback notification or a denial feedback notification.

[0044] In some implementations, detailed logistics dispatch instructions are sent to the selected target carpooling car owners through an interface with the car owner client (such as a mobile application or a vehicle terminal system), requiring the car owner to confirm the order and provide feedback. The car owner can view the detailed task information on the client and click the "Confirm" or "Reject" option after confirmation.

[0045] In step S231, when a confirmation feedback notification from the vehicle owner client regarding the logistics scheduling instruction is detected, a logistics scheduling and delivery task is generated.

[0046] Specifically, all relevant information (vehicle information, route, package information, etc.) will be extracted from the dispatch instruction, a delivery task will be formally generated, and the task details will be transmitted to the vehicle owner. The vehicle owner will be notified of the task confirmation through the vehicle owner's client, including the task number, route, estimated arrival time, etc., and will require the vehicle owner to perform the task according to the specified route and time. At the same time, the delivery status will be updated on the logistics dispatch platform, and real-time tracking will be started to ensure visualization and management of the delivery process.

[0047] In step S233, when a negative feedback notification from the car owner client regarding the logistics scheduling instruction is detected, the remaining rideshare reservation order information, delivery station address and remote package logistics information except the target rideshare reservation order information are input into the logistics scheduling matching model to re-determine the target rideshare reservation order information.

[0048] Specifically, from the initially screened candidate ride-sharing reservations, the rejected target ride-sharing reservation information is removed, and the remaining ride-sharing reservation information, delivery station address, and remote package logistics information are re-input into the logistics scheduling matching model. The logistics scheduling matching model re-optimizes the selection of the remaining ride-sharing vehicles based on factors such as the timeliness of the package, the delivery station address, the ride-sharing route, and the vehicle status, generates a new target ride-sharing reservation, and updates the task list on the car owner's side, and continues to wait for feedback from other car owners.

[0049] Through the real-time feedback mechanism provided by the embodiment of the present invention, after the car owner confirms the order, the system can quickly generate and push the task, thereby reducing the time to start the task and improving the efficiency of the overall logistics scheduling. In addition, when the car owner refuses to accept the order, the scheduling platform can quickly adjust and re-optimize the scheduling plan, avoiding the interruption of the delivery task and ensuring the continuity and efficiency of logistics distribution. Therefore, through dynamic matching and real-time feedback, through the car owner's confirmation of the order and real-time task tracking, both the car owner and the user can get a better service experience, ensuring that the package is delivered on time, while reducing delivery delays caused by scheduling failures.

[0050] Figure 3 An operational flowchart of an example of performing preliminary screening of each ride-sharing reservation order information according to an embodiment of the present invention is shown.

[0051] like Figure 3 As shown, in step S310, for each ride-sharing reservation order information, the package attribute matching degree, time matching degree and space matching degree relative to the ride-sharing reservation order information are calculated according to the package attributes, expected package arrival time, package delivery address and delivery station address, and are combined to obtain the corresponding order matching degree.

[0052] Specifically, based on various information of the package (such as package attributes, timeliness, delivery address, etc.), it is matched with the itinerary, departure time and other data of the ride-sharing car in order to evaluate the suitability of each ride-sharing car for the package delivery needs.

[0053] It should be noted that the matching calculation methods of each dimension can be diverse and should not be limited here. For example, for the parcel attribute matching, the difference value calculation method (such as Euclidean distance, cosine similarity, etc.) can be used for calculation; for time matching, the overlap measure between the travel time of the ride-sharing car and the parcel time window (such as intersection, time difference measure) can be calculated; for spatial matching, spatial matching can be performed based on the geographic information of the parcel delivery address, distribution station address and ride-sharing route, such as calculating Haversine distance, Manhattan distance, etc., to measure the spatial proximity between the delivery route and the ride-sharing route.

[0054] Then, the package attribute matching, time matching and space matching are summarized, such as weighted average or other fusion methods (such as multi-attribute decision-making method), to obtain the comprehensive order matching degree of each ride-sharing and package delivery task.

[0055] In step S320, a preset number of ride-sharing reservation order information with high corresponding order matching rankings are screened from various ride-sharing reservation order information to determine at least one matching candidate ride-sharing reservation order information.

[0056] In some embodiments, the order matching degree is sorted from high to low, and a preset number of top-ranked rideshares (for example, the top 3 or top 5 rideshares) are selected for subsequent dispatch. This number can also be adjusted according to actual needs and business models. Preferably, if some rideshares fail to accept orders due to force majeure (such as failure, owner refusal, etc.), the next candidate rideshare with a high matching degree is automatically selected for re-dispatching.

[0057] Through the embodiments of the present invention, the matching evaluation of package attributes, time, and space is refined, and more accurate scheduling decisions can be made to ensure that the package can choose the most suitable ride-sharing car for delivery. In addition, by sorting and screening the ride-sharing reservation information, the system maximizes the use of the ride-sharing transportation capacity, improves resource allocation efficiency, and reduces the waste of empty trips and unreasonable routes. When faced with different types of packages or a variety of ride-sharing orders, the matching criteria can be flexibly adjusted to adapt to different logistics needs, ensuring the stability and efficiency of the distribution network.

[0058] Regarding the implementation details of the above step S310, in some implementations, it can be implemented in the following ways:

[0059]

[0060] Where M total Indicates the order matching degree, M attr 、M time and M space They represent attribute matching, time matching and space matching respectively, and α1, α2 and α3 represent the weight factors of the corresponding matching dimensions.

[0061] The attribute matching, time matching and space matching are integrated by using the weighted product form and normalized index Ensure that the final matching value is within a reasonable range. Compared with the simple linear weighted sum, the weighted product form can better reflect the synergy between dimensions. For example, if the matching degree of a dimension is 0 (completely mismatched), the total matching degree is immediately 0.

[0062] It should be noted that the weights α1, α2, and α3 can be dynamically adjusted according to business needs or business scenarios. For example, for different business scenarios (such as urgent package delivery), the time matching weight α2 can be increased to give priority to timeliness.

[0063] M attr =ω1·f1(w,W max )+ω2·f2(v,V max )+ω3·f3(t,T support ), Formula (2)

[0064]

[0065]

[0066]

[0067] Where, f1(w,W max )、f2(v,V max ) and f3(t,T support ) represent weight matching function, volume matching function and type matching function respectively, ω1, ω2 and ω3 represent the weights of the corresponding attribute dimensions; w represents the package weight, W max represents the maximum load of the ride-sharing car, v represents the volume of the package, V max represents the maximum volume of the vehicle; t represents the package type, T support Represents the set of types supported by the vehicle.

[0068] Here, the calculation of attribute matching takes into account the weight matching function, volume matching function and type matching function. In the weight matching function and volume matching function, the matching degree is close to the vehicle load limit W when the weight w is close to the vehicle load limit W. max The linear form This avoids the insensitivity of linear descent and ensures that the vehicle can effectively carry the package. In the type matching function, only when the package type t belongs to the type set T supported by the vehicle support When the vehicle is in the cold chain, the matching degree is 1, otherwise it is directly 0, ensuring that the package type (such as cold chain, fragile) is consistent with the vehicle's support capabilities.

[0069] Furthermore, the attribute matching degree flexibly adjusts the importance of different attributes through the weighting mechanism ω1, ω2, ω3. For example, for fragile items, the weight of ω3 can be increased.

[0070]

[0071] Where, T deliver Indicates the expected arrival time of the package, T depart represents the departure time of the ride-sharing car, and k represents the time sensitivity factor.

[0072] Here, an exponential decay function is used to process the time difference T deliver -T depart , reflecting the influence of time nonlinearly. When the vehicle departure time T depart Close to the expected arrival time of the package T deliver When the time difference is large, the matching degree is high; when the time difference is large, the matching degree drops rapidly. The time sensitivity factor k is used to control the time sensitivity. For example, the greater the time urgency, the higher the corresponding k value, making the impact of the time difference on the matching degree more significant.

[0073] M space =f station (L route ,L station )+f delivery (L route ,L package ), Formula (7)

[0074]

[0075]

[0076] In the formula, f station and f delivery They represent the compatibility function of the ride-sharing route with the distribution station address and the package delivery address respectively; L route represents the driving path of the ride-sharing car, L station Indicates the address of the distribution station, L package Indicates the package delivery address; dist(L station ,L route ) represents the distance between the distribution station address and the nearest point on the ride-sharing path, dist(L package ,L route) represents the distance between the package delivery address and the nearest point on the ride-sharing path, and σ1 and σ2 represent the tolerance parameters of the corresponding distances, respectively.

[0077] Here, the exponential decay function is used to process the distance between the distribution station address and the package delivery address and the nearest point of the ride-sharing path. The closer the distance, the higher the matching degree. σ1 and σ2 are used to control the sensitivity of the matching degree to the distance change to meet the configuration requirements of different delivery scenarios. Through the path coverage calculation in the spatial matching degree, the ride-sharing that covers the distribution station and delivery address path is preferentially selected, thereby reducing the additional detour cost.

[0078] Through the embodiment of the present invention, a comprehensive evaluation method of attribute matching, time matching and space matching is introduced to ensure that the matching of packages and ride-sharing is more accurate, significantly reducing the probability of inefficient matching. In addition, by using nonlinear formulas such as exponential decay, the matching degree is made more sensitive under boundary conditions (such as time constraints and large spatial deviations), avoiding the limitations of traditional linear models. In addition, operators can dynamically adjust the weight settings according to actual scenario requirements to achieve intelligent and flexible scheduling and meet the personalized scheduling requirements of different types of remote areas.

[0079] Figure 4 An operational flowchart showing an example of determining target rideshare reservation information from various candidate rideshare reservation information based on a logistics scheduling matching model according to an embodiment of the present invention.

[0080] like Figure 4 As shown, in step S410, a logistics scheduling diagram structure is constructed based on the remote package logistics information and the information of each candidate ride-sharing reservation order.

[0081] Here, the logistics scheduling graph structure includes multiple graph nodes and edge connections, and the multiple graph nodes include package nodes, ride-sharing nodes and global nodes; the node characteristics of the package node are defined based on the remote package logistics information, and the node characteristics of the ride-sharing node are defined based on the candidate ride-sharing reservation order information; the global node is connected to each ride-sharing node for summarizing the global characteristics; the edge weight of the package-ride-sharing edge is defined based on the order matching degree corresponding to the connected node.

[0082] Specifically, the logistics scheduling graph structure consists of a package node, multiple ride-sharing nodes, and a global node. The edge weight of the package-ride-sharing edge determines the strength of the relationship between each package and the ride-sharing. Each graph node in the logistics scheduling graph structure carries a feature vector, and each edge carries a weight value, which indicates the degree of match between the package and the ride-sharing. The role of the global node is to aggregate the feature information of the entire scheduling system. It can summarize the features of all ride-sharing nodes and provide global information for the graph neural network.

[0083] In step S420, the logistics scheduling graph structure is input into the graph neural network to update the node characteristics of each ride-sharing node and the global node according to the package characteristics and edge weights, so as to determine the matching score corresponding to each ride-sharing node.

[0084] It should be noted that graph neural networks are a type of deep learning model specifically designed to process graph data. They can propagate information through the edges between nodes, thereby updating the features of the nodes. In the logistics scheduling graph, the task of the graph neural network is to iteratively update the feature vectors of the nodes based on the package features, ride-sharing node features, and edge weights, so that it can more accurately represent the relationships between the nodes.

[0085] It should be understood that the types of graph neural networks can be diverse, such as graph convolutional networks (GCN) or other types of graph neural networks (such as GraphSAGE, GAT, etc.), which implement information propagation and aggregation. The feature vector of each node will be updated through the features of adjacent nodes to obtain a more meaningful node representation.

[0086] At each layer of the network, nodes update their features based on the information of their neighboring nodes. Through the propagation of multiple graph neural network layers, node features gradually contain more global information and local structural information. After the updated ride-sharing node features are processed, the matching score of each ride-sharing node is calculated based on its final features to reflect the overall matching degree between different ride-sharing and packages.

[0087] In step S430, the target ride-sharing reservation information is determined based on the candidate ride-sharing reservation information corresponding to the ride-sharing node with the highest matching score.

[0088] In some implementations, the ride-sharing reservation information corresponding to the ride-sharing node with the highest matching degree is selected, and the dispatch is confirmed, and then the system outputs the target ride-sharing reservation information, including vehicle information, departure time, route, etc., as the final delivery plan. Then, a specific dispatch instruction is generated, and the relevant car owner client is notified to execute the delivery task.

[0089] Through the embodiments of the present invention, the multiple relationships between packages and rideshares can be fully expressed by utilizing the graph structure. Through information transmission and feature updates, the scheduling accuracy is improved, ensuring that each package can find the most suitable rideshare for delivery. In addition, through the introduction of global nodes and the feature propagation of the graph neural network, optimization can be performed from a global perspective, ensuring the efficiency and stability of the overall logistics scheduling. As a result, the graph neural network can process dynamically updated scheduling information, and the scheduling platform can be flexibly adjusted in the face of different demands and environmental changes to ensure maximum scheduling efficiency.

[0090] Regarding the implementation details of step S420, in some embodiments, the graph neural network adopts a heterogeneous graph neural network, which includes a cascaded message passing mechanism module, an edge feature update module and an output module.

[0091] Here, a heterogeneous graph neural network (Hetero-GNN) is used, which contains package nodes, ride-sharing nodes, and global nodes, and is suitable for describing the complex relationships between different types of objects (packages, vehicles, system status). Based on layer-by-layer message passing, nodes pass messages through edges and update features, aggregate neighborhood information layer by layer, and finally obtain a global representation of each node.

[0092] The graph neural network updates node embeddings through multiple rounds of message passing. The message passing mechanism module is used to perform the following operations:

[0093] Message passing from the package node to the ride-sharing node:

[0094]

[0095]

[0096]

[0097] Where l represents the layer index of the graph neural network, is the feature vector of the wrapped node p in layer l, and are the feature vectors output by the ride-sharing node c at the lth layer and the l+1th layer respectively, is the representation of the message transmitted from the package node p to the ride-sharing node c at the l+1 layer; is the representation of the edge weight between the parcel node p and the ride-sharing node c at the lth layer, which is defined by the corresponding order matching degree; and are the weight matrices of linear transformation of package features, linear transformation of ride-sharing features and linear transformation of edge features respectively, ReLU is the ReLU nonlinear activation function; N(c) is the set of neighboring package nodes of ride-sharing node c, is the feature vector of the neighbor wrapping node q at layer l; is the representation of the attention weights of the package node p and the ride-sharing node c at the l+1 layer, a represents the attention weight vector, and ‖ represents the vector concatenation operation.

[0098] Here, the parcel node transmits the logistics characteristics (such as urgency, weight, volume, etc.) to the ride-sharing node. It is the interaction description between the package and the ride-sharing, combined with the edge features Dynamically reflect the current matching relationship.

[0099] By adopting a dynamic edge feature update mechanism, the edge weights are adjusted in real time according to the changes in the characteristics of the package and ride-sharing nodes, and factors such as path deviation distance and time window adaptation are reflected in real time. The personalized matching relationship between different packages and ride-sharing is captured, ensuring that the matching of packages and ride-sharing is more accurate and reducing the waste of resources caused by static rules.

[0100] Attention Weight Based on the characteristics of packages and ride-sharing, the model can distinguish the importance of different packages and give priority to the delivery needs of high-timeliness packages, while making full use of ride-sharing resources and balancing timeliness and resource utilization.

[0101] Pass messages from the rideshare node to the global node to aggregate overall resource and timeliness information:

[0102]

[0103] In the formula, is the representation of the message from the ride-sharing node c to the global node g at the l+1 layer, is the weight matrix transferred from the ride-sharing node to the global node; and are the feature vectors of the global node g at the l+2 layer and the l+1 layer respectively, N(g) represents the set of ride-sharing nodes connected to the global node g, A weight matrix representing global node feature updates.

[0104] Here, the ride-sharing node reports its comprehensive characteristics to the global node g, reflecting the current state of transportation resource utilization. The design of the global node introduces system-level optimization information into the node characteristics, enhancing the global perspective of the scheduling model.

[0105] The global node g sends the updated global information through the message Passed to each ride node c:

[0106]

[0107]

[0108] In the formula, It represents the message from the global node g to the ride-sharing node c at the l+2 layer, represents the weight matrix transferred from the global node to the ride-sharing node, is the feature vector output by the ride-sharing node c at the l+3 layer.

[0109] Here, the global node feeds back the system optimization information to the ride-sharing node, guides the feature adjustment of the ride-sharing node, and optimizes the overall scheduling strategy. The global node summarizes the status information of all ride-sharing nodes, guides the scheduling strategy of the ride-sharing node through the feedback mechanism, and optimizes the global resource allocation while meeting the needs of a single package, avoiding local optimization and global imbalance problems.

[0110] The edge feature update module is used to update the features of the package-ride edge after each layer of message passing, so that it can be dynamically adjusted to adapt to changes in node features:

[0111]

[0112] In the formula, is the representation of the edge weight between the parcel node p and the ride-sharing node c at the l+1 layer; MLP e It is a multi-layer perceptron, which is used to dynamically adjust the nonlinear mapping function of edge features.

[0113] The feature fusion of parcel nodes and ride-sharing nodes is considered through the multi-layer perceptron (MLP), and the edge features are dynamically adjusted so that the edge features can reflect the changes in node features, and the changes in edge features can adapt to the real-time adjustment of node features.

[0114] The output module is used to calculate the matching score based on the final node features of the ride-sharing node:

[0115]

[0116] In the formula, S c represents the matching score of the ride-sharing node c, W o and b o Represent the weight matrix and bias term of the output layer, σ o is the activation function of the output layer, L is the depth of the graph neural network, It is the feature vector output by the hitchhiking node c at the final Lth layer.

[0117] Here, the ride-sharing node After multiple layers of message transmission, it not only integrates the characteristics of adjacent packages, but also combines the global system optimization information, has a global vision, and can achieve more reasonable path planning and package allocation in actual scheduling.

[0118] In the graph neural network provided by the embodiment of the present invention, each layer of message transmission and feature update is based on the neighborhood of nodes and edges, and the computational complexity is low. In large-scale logistics scheduling scenarios, scheduling decisions can be generated quickly to meet real-time requirements. In addition, by dynamically updating edge weights, repeated calculations of full-graph relationships are avoided, and only local adjustments are required, reducing computational costs and keeping computing resource consumption controllable.

[0119] Through the embodiments of the present invention, based on the characteristics of the parcel node and the ride-sharing node, through layer-by-layer information aggregation and dynamic edge feature update, it is possible to accurately evaluate the real-time matching degree of each ride-sharing node, dynamically adapt to the real-time changes in the parcel demand and ride-sharing resource status in the logistics scenario, and improve the matching efficiency of the parcel and ride-sharing resources. In addition, by introducing global nodes, the heterogeneous graph neural network can comprehensively consider the system-level resource distribution, realize the optimization feedback of the ride-sharing node, improve the overall resource utilization of the system, and avoid the defects of local optimization caused by ignoring the global perspective.

[0120] In some examples of the embodiments of the present invention, the heterogeneous graph neural network is optimized and trained based on a data sample set and a loss function. Specifically, the loss function of the heterogeneous graph neural network is:

[0121]

[0122]

[0123] L time (x i )=max(0,T i deliver -T i real ), Formula (21)

[0124]

[0125] Where L represents the total loss function of the heterogeneous graph neural network, N is the total number of samples in the data sample set, i is the sample index, and x i is the input feature of sample i, y i is the true label of sample i; L match (x i ,y i )、L time (x i ) and L space (x i ) represent the matching loss, time-sensitive loss and spatial constraint loss of sample i respectively, λ1, λ2, λ3 are the weight parameters corresponding to each loss item; S i represents the matching score for sample i output by the model; T i deliver represents the estimated arrival time of the package in sample i, T i real represents the actual arrival time of the package in sample i; P i represents the set of packages assigned to rideshare c(i) in sample i, L c(i) represents the current position of the ride-sharing car in sample i, Lp represents the delivery address of package p in sample i, dist(L c(i) ,L p ) indicates L c(i) With L p The path distance between c(i) ,L station ) indicates L c(i) With L station The path distance between them.

[0126] In an embodiment of the present invention, the loss function of the heterogeneous graph neural network integrates matching loss, time-sensitive loss and spatial constraint loss.

[0127] First, in the logistics scheduling scenario, accurately matching packages and ride-sharing nodes is the most basic optimization goal. The matching score S predicted by the model is measured by the matching loss. i and the true label y i The consistency of the model is calculated by using the cross entropy function to quantify the accuracy of the model's judgment on matches and non-matches.

[0128] Secondly, in the logistics and distribution scenario, time sensitivity is also an important optimization goal. Through time-sensitive losses, the delivery tasks of urgent packages can be prioritized to ensure that they can be delivered within the specified time window. Specifically, a timeout penalty mechanism is designed, and through the max(0,·) function, additional losses are generated when the estimated arrival time exceeds the actual arrival time. In addition, the time sensitivity of each sample is calculated independently, and urgent orders are automatically prioritized in model training, which can dynamically optimize the scheduling plan according to the urgency of the time window.

[0129] Third, in the logistics and distribution scenario, optimizing the path is a key link in controlling costs and improving efficiency. By introducing spatial constraint loss, the path planning of the ride-sharing service is constrained, the path deviation is minimized, and excessive invalid driving distance is avoided, thereby improving resource utilization. Specifically, by calculating the deviation ratio of the distribution path, the path planning of the ride-sharing service can be dynamically adjusted to shorten the distribution path as much as possible. The spatial constraint loss enables the model to not only focus on matching accuracy, but also optimize the path consumption in actual transportation.

[0130] Through the embodiments of the present invention, the total loss function of the heterogeneous graph neural network integrates the multi-objective constraints of matching accuracy, time sensitivity and path planning. By dynamically adjusting the weights through λ1, λ2, and λ3, it is possible to simultaneously optimize accuracy, timeliness and path planning, achieve comprehensive performance optimization in different scenarios, and meet the actual needs of various logistics scheduling scenarios.

[0131] The following is a description of the logistics scheduling system for e-commerce trade provided by the present invention. The logistics scheduling system for e-commerce trade described below and the logistics scheduling method for e-commerce trade described above can be referenced to each other.

[0132] Figure 5 A structural block diagram of an example of a logistics scheduling system for e-commerce trade according to an embodiment of the present invention is shown.

[0133] like Figure 5 As shown, the logistics scheduling system 500 for e-commerce trade includes a first acquisition unit 510, a second acquisition unit 520, an appointment form initial screening unit 530, a scheduling matching unit 540 and a scheduling execution unit 550.

[0134] The first acquisition unit 510 is used to acquire remote package logistics information, where the remote package logistics information includes a package delivery address, package attributes, time urgency, and expected arrival time of the package in a corresponding remote area.

[0135] The second acquisition unit 520 is used to obtain at least one ride-sharing reservation order information corresponding to the travel destination of the remote area, and the ride-sharing reservation order information includes basic vehicle information, driving route, departure time and logistics scheduling historical order records.

[0136] The reservation order initial screening unit 530 is used to initially screen each of the ride-sharing reservation order information based on the package attributes, the expected arrival time of the package, the package delivery address and the distribution station address to determine at least one matching candidate ride-sharing reservation order information.

[0137] The dispatch matching unit 540 is used to input each of the candidate ride-sharing reservation order information and the remote package logistics information into the logistics dispatch matching model to determine the target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information.

[0138] The scheduling execution unit 550 is used to update the driving route in the target ride-sharing reservation order information according to the distribution station address and the package delivery address, and send logistics scheduling instructions to the corresponding car owner client.

[0139] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of combined actions, but those skilled in the art should be aware that the present invention is not limited by the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0140] In some embodiments, an embodiment of the present invention provides a non-volatile computer-readable storage medium, which stores one or more programs including execution instructions, and the execution instructions can be read and executed by electronic devices (including but not limited to computers, servers, or network devices, etc.) to execute the above-mentioned logistics scheduling method for e-commerce trade of the present invention.

[0141] In some embodiments, an embodiment of the present invention further provides a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the above-mentioned logistics scheduling method for e-commerce trade.

[0142] In some embodiments, an embodiment of the present invention also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute a logistics scheduling method for e-commerce trade.

[0143] Figure 6 FIG. 1 is a schematic diagram of the hardware structure of an electronic device for executing a logistics scheduling method for e-commerce trade provided by another embodiment of the present invention. Figure 6 As shown, the device includes:

[0144] One or more processors 610 and memory 620, Figure 6 A processor 610 is taken as an example.

[0145] The device for executing the logistics scheduling method for electronic commerce trade may further include: an input device 630 and an output device 640 .

[0146] The processor 610, the memory 620, the input device 630 and the output device 640 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0147] The memory 620 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the logistics scheduling method for e-commerce trade in the embodiment of the present invention. The processor 610 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 620, that is, the logistics scheduling method for e-commerce trade in the above method embodiment is implemented.

[0148] The memory 620 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 620 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 620 may optionally include a memory remotely arranged relative to the processor 610, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0149] The input device 630 can receive input digital or character information and generate signals related to user settings and function control of the electronic device. The output device 640 can include a display device such as a display screen.

[0150] The one or more modules are stored in the memory 620, and when executed by the one or more processors 610, the logistics scheduling method for e-commerce trade in any of the above method embodiments is executed.

[0151] The above product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0152] The electronic devices of the embodiments of the present invention exist in various forms, including but not limited to:

[0153] (1) Mobile communication equipment: This type of equipment is characterized by having mobile communication functions and its main purpose is to provide voice and data communications. This type of terminal includes: smart phones, multimedia phones, functional phones, and low-end phones.

[0154] (2) Ultra-mobile personal computer devices: These devices belong to the category of personal computers, have computing and processing functions, and generally also have mobile Internet access features. These terminals include: PDA, MID and UMPC devices, etc.

[0155] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players, handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0156] (4) Other onboard electronic devices with data interaction functions, such as on-board devices installed in vehicles.

[0157] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A logistics scheduling method for e-commerce trade, comprising: Obtaining remote package logistics information, wherein the remote package logistics information includes a package delivery address, package attributes, time urgency, and expected arrival time of the package in the corresponding remote area; Acquire at least one ride-sharing reservation order information corresponding to the trip destination of the remote area, wherein the ride-sharing reservation order information includes basic vehicle information, travel route, departure time, and logistics scheduling historical order records; Preliminarily screening each of the ride-sharing reservation order information by using the package attributes, the expected arrival time of the package, the package delivery address and the delivery station address to determine at least one matching candidate ride-sharing reservation order information; Inputting each of the candidate ride-sharing reservation order information and the remote package logistics information into a logistics scheduling matching model to determine the target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information; The driving route in the target ride-sharing reservation order information is updated according to the distribution station address and the package delivery address, and the logistics dispatch instruction is sent to the corresponding car owner client.

2. The method according to claim 1, wherein: After sending the logistics dispatch instruction to the corresponding vehicle owner client, the method further includes: When a confirmation feedback notification of the vehicle owner client to the logistics dispatch instruction is detected, a logistics dispatch delivery task is generated; When a negative feedback notification from the vehicle owner client regarding the logistics scheduling instruction is detected, the remaining rideshare reservation order information, the delivery station address and the remote package logistics information except the target rideshare reservation order information are input into the logistics scheduling matching model to re-determine the target rideshare reservation order information.

3. The method according to claim 1, wherein: The preliminarily screening each of the ride-sharing reservation order information by using the package attributes, the expected arrival time of the package, the package delivery address and the delivery station address to determine at least one matching candidate ride-sharing reservation order information includes: For each of the ride-sharing reservation order information, according to the package attributes, the expected arrival time of the package, the package delivery address and the delivery station address, calculate the package attribute matching degree, time matching degree and space matching degree relative to the ride-sharing reservation order information, and combine them to obtain the corresponding order matching degree; A preset number of ride-sharing reservation order information with high order matching rankings are screened from each of the ride-sharing reservation order information to determine at least one matching candidate ride-sharing reservation order information.

4. The method according to claim 3, wherein: For each of the ride-sharing reservation information, according to the package attributes, the expected arrival time of the package, the package delivery address and the delivery station address, the package attribute matching degree, time matching degree and space matching degree relative to the ride-sharing reservation information are calculated, and the corresponding order matching degree is obtained by combining them, including: Where M total Indicates the order matching degree, M attr 、M time and M space They represent attribute matching, time matching and space matching respectively, and α1, α2 and α3 represent the weight factors of the corresponding matching dimensions; M attr =ω1·f1(w,W max )+ω2·f2(v,V max )+ω3·f3(t,T support ), Where, f1(w,W max )、f2(v,V max ) and f3(t,T support ) represent weight matching function, volume matching function and type matching function respectively, ω1, ω2 and ω3 represent the weights of the corresponding attribute dimensions; w represents the package weight, W max represents the maximum load of the ride-sharing car, v represents the volume of the package, V max represents the maximum volume of the vehicle; t represents the package type, T support Represents the type set supported by the vehicle; Where, T deliver Indicates the expected arrival time of the package, T depart represents the departure time of the ride-sharing car, and k represents the time sensitivity factor; M space =f station (L route ,L station )+f delivery (L route ,L package ), In the formula, f station and f delivery They represent the compatibility function of the ride-sharing route with the distribution station address and the package delivery address respectively; L route represents the driving path of the ride-sharing car, L station Indicates the address of the distribution station, L package Indicates the package delivery address; dist(L station ,L route ) represents the distance between the distribution station address and the nearest point on the ride-sharing path, dist(L package ,L route ) represents the distance between the package delivery address and the nearest point on the ride-sharing path, and σ1 and σ2 represent the tolerance parameters of the corresponding distances, respectively.

5. The method according to claim 3, wherein: Inputting each of the candidate ride-sharing reservation order information, the distribution station address and the remote package logistics information into a logistics scheduling matching model to determine the target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information, including: Based on the remote parcel logistics information and the information of each candidate ride-sharing reservation order, a logistics scheduling graph structure is constructed; the logistics scheduling graph structure includes multiple graph nodes and edge connections, and the multiple graph nodes include parcel nodes, ride-sharing nodes and global nodes; the node characteristics of the parcel nodes are defined according to the remote parcel logistics information, and the node characteristics of the ride-sharing nodes are defined according to the candidate ride-sharing reservation order information; the global node is connected to each of the ride-sharing nodes for summarizing global characteristics; the edge weight of the parcel-ride-sharing edge is defined according to the order matching degree corresponding to the connected node; Inputting the logistics scheduling graph structure into the graph neural network to update the node features of each of the ride-sharing nodes and the global node according to the package features and edge weights, thereby determining the matching score corresponding to each ride-sharing node; The target ride-sharing reservation information is determined based on the candidate ride-sharing reservation information corresponding to the ride-sharing node with the highest matching score.

6. The method according to claim 5, wherein: The graph neural network adopts a heterogeneous graph neural network, and the heterogeneous graph neural network includes a cascaded message passing mechanism module, an edge feature update module and an output module; The message delivery mechanism module is used to perform the following operations: Message passing from the package node to the ride-sharing node: Where l represents the layer index of the graph neural network, is the feature vector of the wrapped node p in layer l, and are the feature vectors output by the ride-sharing node c at the lth layer and the l+1th layer respectively, is the representation of the message transmitted from the package node p to the ride-sharing node c at the l+1 layer; is the representation of the edge weight between the parcel node p and the ride-sharing node c at the lth layer, which is defined by the corresponding order matching degree; and are the weight matrices of linear transformation of package features, linear transformation of ride-sharing features and linear transformation of edge features respectively, ReLU is the ReLU nonlinear activation function; N(c) is the set of neighboring package nodes of ride-sharing node c, is the feature vector of the neighbor wrapping node q at layer l; is the representation of the attention weights of the parcel node p and the ride-sharing node c at the l+1 layer, a represents the attention weight vector, ‖ represents the vector concatenation operation; Pass messages from the rideshare node to the global node to aggregate overall resource and timeliness information: In the formula, is the representation of the message from the ride-sharing node c to the global node g at the l+1 layer, is the weight matrix transferred from the ride-sharing node to the global node; and are the feature vectors of the global node g at the l+2 layer and the l+1 layer respectively, N(g) represents the set of ride-sharing nodes connected to the global node g, The weight matrix representing the global node feature update; The global node g sends the updated global information through the message Passed to each ride node c: In the formula, Represents the message from the global node g to the ride-sharing node c at the l+2 layer, represents the weight matrix transferred from the global node to the ride-sharing node, is the feature vector output by the ride-sharing node c at the l+3 layer; The edge feature update module is used to update the features of the package-ride edge after each layer of message transmission, so that it can be dynamically adjusted to adapt to the changes in node features: In the formula, is the representation of the edge weight between the parcel node p and the ride-sharing node c at the l+1 layer; MLP e It is a multi-layer perceptron, which is used to dynamically adjust the nonlinear mapping function of edge features; The output module is used to calculate the matching score according to the final node features of the ride-sharing node: In the formula, S c represents the matching score of the ride-sharing node c, W o and b o Represent the weight matrix and bias term of the output layer, σ o is the activation function of the output layer, L is the depth of the graph neural network, It is the feature vector output by the hitchhiking node c at the final Lth layer.

7. The method according to claim 6, wherein: The loss function of the heterogeneous graph neural network is: L match (x i ,y i )=-(y i log(S i )+(1-y i )log(1-S i )), L time (x i )=max(0,T i deliver -T i real ), Where L represents the total loss function of the heterogeneous graph neural network, N is the total number of samples in the data sample set, i is the sample index, and x i is the input feature of sample i, y i is the true label of sample i; L match (x i ,y i )、L time (x i ) and L space (x i ) represent the matching loss, time-sensitive loss and spatial constraint loss of sample i respectively, λ1, λ2, λ3 are the weight parameters corresponding to each loss item; S i represents the matching score for sample i output by the model; T i deliver represents the estimated arrival time of the package in sample i, T i real represents the actual arrival time of the package in sample i; P i represents the set of packages assigned to rideshare c(i) in sample i, L c(i) represents the current position of the ride-sharing car in sample i, L p represents the delivery address of package p in sample i, dist(L c(i) ,L p ) indicates L c(i) With L p The path distance between c(i) ,L station ) indicates L c(i) With L station The path distance between them.

8. A logistics dispatching system for e-commerce trade, comprising: A first acquisition unit is used to acquire remote package logistics information, wherein the remote package logistics information includes a package delivery address, package attributes, time urgency, and expected arrival time of the package in a corresponding remote area; A second acquisition unit is used to acquire at least one ride-sharing reservation order information corresponding to a trip destination of the remote area, wherein the ride-sharing reservation order information includes basic vehicle information, driving route, departure time, and logistics scheduling history order records; A reservation order initial screening unit, used to initially screen each of the ride-sharing reservation order information based on the package attributes, the expected arrival time of the package, the package delivery address and the delivery station address, so as to determine at least one matching candidate ride-sharing reservation order information; A dispatching and matching unit, used for inputting each of the candidate ride-sharing reservation order information and the remote package logistics information into a logistics dispatching and matching model, so as to determine the target ride-sharing reservation order information from each of the candidate ride-sharing reservation order information; The scheduling execution unit is used to update the driving route in the target ride-sharing reservation order information according to the distribution station address and the package delivery address, and send logistics scheduling instructions to the corresponding car owner client.

Images