Methods, apparatus, equipment and computer-readable media for delivering goods

Abstract

This invention discloses a method, apparatus, equipment, and computer-readable medium for delivering goods, relating to the field of logistics and transportation technology. One specific embodiment of the method includes: aggregating historical orders into multiple order clusters based on order region and order processing time; inputting the order clusters, order processing time, and order regions into a path model, whereby the path model outputs historical delivery routes and historical connection points of the delivery regions, wherein the historical delivery routes and historical connection points satisfy the order processing time of the historical orders; inputting the order region and order processing time of the current order into the path model, whereby the path model outputs the current delivery route and the current connection point of the delivery region; and delivering the goods of the current order using a current connection route constructed from the current delivery route and the current connection point. This embodiment can accelerate the delivery speed of goods and ensure timely delivery.

Description

Technical Field

[0001] This invention relates to the field of logistics and transportation technology, and in particular to a method, apparatus, equipment, and computer-readable medium for distributing goods. Background Technology

[0002] Driven by e-commerce, logistics companies have developed rapidly, and cities have built corresponding express delivery systems. Generally, logistics companies divide cities into districts according to administrative boundaries and road networks. The picking and distribution of packages in each district are handled by specific business departments. Each business department further subdivides the area into multiple delivery zones based on the building characteristics within its service area, and arranges delivery personnel to handle the pickup and delivery tasks for each delivery zone.

[0003] In developing this invention, the inventors discovered at least the following problems in the prior art: some delivery areas are far from the sales office, and the delivery end spends a lot of time en route, thus requiring shuttle buses to transport packages between the sales office and each delivery area. The delivery end waits for the shuttle buses to deliver packages, resulting in long delivery times. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, device, and computer-readable medium for delivering goods, which can speed up the delivery of goods and ensure the delivery time limit.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method for delivering articles is provided, comprising:

[0006] Based on the order region and order processing time, historical orders are aggregated into multiple order clusters.

[0007] The order clustering region, the order processing time, and the order region are input into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order.

[0008] The order region and order processing time of the current order are input into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point.

[0009] Based on order region and order processing time, historical orders are aggregated into multiple order cluster regions, including:

[0010] Based on the order processing time, the historical orders are divided into orders in multiple time periods;

[0011] Based on the order region, orders within the same time period are aggregated to obtain the multiple order cluster regions.

[0012] The step of inputting the order clustering region, the order processing time, and the order region into the path model, wherein the path model outputs the historical delivery route of the delivery end and the historical pick-up point of the delivery region, and the historical delivery route and the historical pick-up point satisfy the order processing time of the historical order, includes:

[0013] The order clustering region, the order processing time, and the order region are input into the path model, and the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region.

[0014] The historical pick-up points in the delivery area meet the pick-up times of the historical orders;

[0015] The actual processing time of the historical orders is determined in the historical delivery route to meet the order processing time requirements of the historical orders.

[0016] The historical pick-up points in the delivery area that meet the pick-up times of the historical orders include:

[0017] The historical pick-up points in the delivery area meet the pick-up times of the historical orders in multiple historical delivery routes.

[0018] The step of inputting the order clustering region, the order processing time, and the order region into the path model, and the path model outputting the historical delivery route of the delivery end and the historical connection point of the delivery region, includes:

[0019] Input the order clustering region, the order processing time, and the order region into the path model;

[0020] The path model outputs the historical delivery path of a delivery terminal and the historical connection point of the corresponding delivery area. After hiding the order area in the delivery area, the path model outputs the historical delivery path of the next delivery terminal and the historical connection point of the corresponding delivery area of ​​the next delivery terminal.

[0021] The step of inputting the order clustering region, the order processing time, and the order region into the path model, and the path model outputting the historical delivery route of the delivery end and the historical connection point of the delivery region, includes:

[0022] The order clustering region, the order processing time, and the order region are input into the path model. After training the path model with a gradient measurement algorithm, the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region.

[0023] The current delivery route and the current connection point together form the current connection route for delivering the items of the current order, including:

[0024] The delivery terminal retrieves the items for the current order at the current pick-up point according to the pick-up time;

[0025] The delivery terminal delivers the items of the current order according to the current delivery route to meet the order processing time.

[0026] According to a second aspect of the present invention, an apparatus for delivering articles is provided, comprising:

[0027] The aggregation module is used to aggregate historical orders into multiple order clusters based on order region and order processing time;

[0028] The training module is used to input the order clustering region, the order processing time, and the order region into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order.

[0029] The output module is used to input the order region and order processing time of the current order into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point.

[0030] According to a third aspect of the present invention, an electronic device for delivering articles is provided, comprising:

[0031] One or more processors;

[0032] Storage device for storing one or more programs.

[0033] When the one or more programs are executed by the one or more processors, the one or more processors perform the methods described above.

[0034] According to a fourth aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method as described above.

[0035] One embodiment of the above invention has the following advantages or beneficial effects: based on the order region and order processing time, historical orders are aggregated into multiple order cluster regions; the order cluster regions, the order processing time, and the order regions are input into a path model, and the path model outputs the historical delivery routes of the delivery end and the historical connection points of the delivery regions, wherein the historical delivery routes and the historical connection points satisfy the order processing time of the historical orders; the order region and order processing time of the current order are input into the path model, and the path model outputs the current delivery route of the delivery end and the current connection point of the delivery region, and the items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point. Using a path model allows for direct knowledge of the delivery route and connection route, thus accelerating the delivery speed and ensuring timely delivery.

[0036] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0037] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0038] Figure 1 This is a schematic diagram of the main flow of a method for delivering items according to an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the process of aggregating historical orders into multiple order clustering regions according to an embodiment of the present invention;

[0040] Figure 3 This is a flowchart illustrating the training path model according to an embodiment of the present invention;

[0041] Figure 4 This is a flowchart illustrating another training path model according to an embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the structure of the training path model according to an embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram illustrating the process of delivering items for the current order according to an embodiment of the present invention;

[0044] Figure 7 This is a schematic diagram of the main structure of a device for delivering goods according to an embodiment of the present invention;

[0045] Figure 8 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied;

[0046] Figure 9This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0047] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0048] The delivery end, responsible for delivering goods within a designated delivery area, significantly impacts the overall efficiency of logistics. Within a delivery area, a large number of logistics orders need to be processed collaboratively, including order allocation, determining the processing order, and selecting connection points with other delivery services. However, the sheer volume of orders and time pressures often lead to low efficiency on the delivery end.

[0049] With the development of database technology and data-driven methods, logistics companies increasingly hope to manage delivery and order scheduling through intelligent means. For example, they might model delivery scheduling as a vehicle routing problem (VRP) and solve it using exact methods, heuristic methods, or deep learning methods. However, considering the complexity of logistics systems, the large order demand, and the nondeterministic polynomial (NP-hard) nature of the vehicle routing problem itself, the following significant limitations exist:

[0050] In collaborative work mode, it is necessary to quickly generate an operation plan based on the current task. However, precise methods, such as branch and bound, are slow in solving problems and cannot meet real-time requirements.

[0051] When using heuristic methods to solve vehicle routing problems, the algorithm iteratively generates solutions based on heuristic operators. However, due to the large scale of the problem, heuristic methods are prone to getting trapped in local optima, failing to guarantee overall optimization performance. Moreover, heuristic methods often only generate scheduling plans based on the tasks of the current day, unable to learn from historical data and predict task patterns, thus making it difficult to effectively utilize data-driven optimization.

[0052] Therefore, there is a waiting period at the delivery end for the receiving end to transmit the items in the order, resulting in a longer delivery time.

[0053] To address the issue of long delivery times, the following technical solutions from the embodiments of the present invention can be adopted.

[0054] See Figure 1 , Figure 1This is a schematic diagram of the main flow of a method for delivering goods according to an embodiment of the present invention. The delivery route and pick-up points are input through a path model to deliver goods according to the order processing time. Figure 1 As shown in Figure 100, the specific steps include:

[0055] S101. Based on the order region and order processing time, aggregate historical orders into multiple order cluster regions.

[0056] In an embodiment of the invention, the items in the order are delivered by a delivery terminal. As an example, the delivery terminal includes a robot that carries the items in the order and delivers them to the delivery address specified in the order.

[0057] A delivery terminal is set up within a delivery area, and that terminal is responsible for orders within that area. As an example, one delivery area corresponds to one delivery terminal.

[0058] The delivery area includes multiple addresses, and each address can be divided into a corresponding order area according to latitude and longitude. An order area is a set of delivery addresses within a preset range. As an example, the set of delivery addresses within 1 kilometer of a preset origin is designated as order area 1. The delivery area includes multiple order areas. Each order area includes multiple delivery addresses.

[0059] Given the considerable distance between the delivery point and the relevant branch office, a connecting terminal is needed to transport the items from the order at a designated pick-up point. As an example, the picking-up point includes a shuttle vehicle. The delivery point retrieves the items from the order at the pick-up point and then transports them to the delivery address. The delivery point must wait for the picking-up point to retrieve the items.

[0060] In real-world applications, orders need to be shipped according to their processing time. For example, order 1 needs to be delivered to the address before 3 PM the following day.

[0061] Considering that delivered items need to meet order processing time requirements, historical orders can be aggregated into multiple order clusters based on order region and order processing time. An order cluster is a collection of order regions established based on order processing time.

[0062] See Figure 2 That is, 200. Figure 2 This is a schematic diagram illustrating the process of aggregating historical orders into multiple order clustering regions according to an embodiment of the present invention. Specifically, it includes the following steps:

[0063] S201. Based on the order processing time, divide historical orders into orders in multiple time periods.

[0064] Each order has a corresponding order processing time. Historical orders can be divided into multiple time periods based on their processing time. For example, order processing times could include: before 9:00 AM; 9:00 AM to 3:00 PM; and 3:00 PM to 8:00 PM, thus dividing historical orders into three time periods.

[0065] S202. Based on the order region, aggregate orders within the same time period to obtain multiple order cluster regions.

[0066] For each order, the order region is determined based on the delivery address. Based on the order region, orders from multiple time periods are aggregated to obtain multiple order clusters. The time window of each order cluster is the time period corresponding to the order.

[0067] exist Figure 2 In one embodiment, historical orders with the same order region and order processing time are aggregated into an order cluster region.

[0068] In one embodiment of the invention, the pick-up point is located in the order area of ​​the delivery area. As an example, one pick-up point is set up in the delivery area. Considering that order clustering areas involve order processing time, from a time perspective, each time period has a corresponding pick-up point. The order areas of pick-up points in multiple time periods can be the same or different.

[0069] S102. Input the order clustering area, order processing time and order area into the path model. The path model outputs the historical delivery path of the delivery end and the historical connection point of the delivery area. The historical delivery path and historical connection point meet the order processing time of the historical order.

[0070] In embodiments of the present invention, delivery routes and connection routes are predicted using a path model. Specifically, the path model can be trained using historical orders. It should be noted that orders have order processing times, meaning that order delivery must meet these times. As an example, the order processing time includes delivery before 3 PM. Therefore, meeting the order processing time can be used as a convergence condition for training the path model. For example, a convergence condition could include meeting the order processing time for 90% of orders.

[0071] The order clustering region, order processing time, and order region are used as input parameters for the path model. The output parameters of the path model include: historical delivery routes at the delivery end and historical pick-up points within the delivery region. The model is trained to obtain the path model when the historical delivery routes and historical pick-up points meet the historical order processing times.

[0072] See Figure 3 That is, 300. Figure 3 This is a flowchart illustrating the training path model according to an embodiment of the present invention. Specifically, it includes the following steps:

[0073] S301. Input the order clustering area, order processing time and order area into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery area.

[0074] Order clustering regions represent areas of orders with the same processing time. The order clustering regions, order processing times, and order regions are used as inputs to the routing model. The routing model outputs the historical delivery routes and historical connection points within the delivery regions.

[0075] In one embodiment of the invention, the path model includes an encoder and a decoder. The encoder is based on an attention mechanism. The decoder is based on both multi-head and single-head attention mechanisms.

[0076] As an example, after the path model outputs the historical delivery path, it determines whether the historical delivery path meets the order processing time requirement based on the order processing time of each order area. The historical delivery path includes the order processing time before 9:00 AM, for order areas 1, 3, 4, and 5. That is, the delivery end starts delivering items from order area 1, then sequentially through order areas 3, 4, and 5, delivering items in each of these order areas to ensure that item delivery is completed before 9:00 AM. Order area 1 serves as the historical connection point for the delivery areas.

[0077] S302, The historical pick-up points in the delivery area meet the pick-up time requirements of historical orders.

[0078] In embodiments of the present invention, the historical delivery route must not only meet the order processing time but also the historical order pick-up time. Pick-up time is the time it takes for the delivery end to retrieve the items from the order at the pick-up point. The pick-up end needs to transport the items from the order to the pick-up point according to the pick-up time.

[0079] From the perspective of the connecting end, the connecting end moves from one connecting point to another according to the connecting time of each connecting point, realizing the transportation of items in the order across multiple connecting points. The connecting point is the order area in the delivery route. In this way, after the delivery end picks up the items in the order at the connecting point, it can directly deliver the items, thereby improving the efficiency of item delivery.

[0080] In one embodiment of the present invention, the historical pick-up points in the delivery area meet the pick-up times of historical orders in multiple historical delivery routes.

[0081] For delivery zones, pick-up points are set up. As an example, a delivery zone typically has one pick-up point. Each delivery zone has a corresponding pick-up point. The pick-up point transports items from an order, and the pick-up time must be met. Therefore, each delivery zone's delivery end corresponds to a delivery route for that zone, resulting in multiple delivery routes for multiple delivery zones. The historical pick-up points of a delivery zone must satisfy the pick-up times of historical orders across multiple historical delivery routes. In other words, the pick-up time needs to be matched across multiple historical delivery routes to enable item pick-up across multiple delivery zones.

[0082] S303. Determine the actual processing time of historical orders in the historical delivery routes to meet the order processing time requirements of historical orders.

[0083] In embodiments of the present invention, a path model can be used to output multiple historical delivery routes and historical connection points within delivery areas. The order processing time of historical orders is used as the convergence condition for training the path model.

[0084] If the historical delivery routes and historical connection points of the delivery area output by the path model meet the order processing time of the historical orders, then the path model has completed training; if the historical delivery routes and historical connection points of the delivery area output by the path model do not meet the order processing time of the historical orders, then the path model continues training.

[0085] exist Figure 3 In one embodiment, the path model is trained using order processing time as the convergence condition.

[0086] See Figure 4 That is, 400. Figure 4 This is a flowchart illustrating another training path model according to an embodiment of the present invention. Specifically, it includes the following steps:

[0087] S401. Input the order clustering region, order processing time, and order region into the path model.

[0088] To train the path model, order clustering regions, order processing times, and order regions can be used as training data and output to the path model.

[0089] In one embodiment of the present invention, in order to improve the accuracy of the path model, the training data can be amplified by adding noise based on the above-mentioned training data.

[0090] S402. The path model outputs the historical delivery path of a delivery terminal and the historical connection point of the corresponding delivery area of ​​the delivery terminal. After hiding the order area in the delivery area, the path model outputs the historical delivery path of the next delivery terminal and the historical connection point of the corresponding delivery area of ​​the next delivery terminal.

[0091] In embodiments of the present invention, the path model needs to output the historical delivery paths of multiple delivery terminals and the historical pick-up points of the corresponding delivery areas of the delivery terminals, and construct the pick-up paths using the historical pick-up points of multiple delivery areas. Here, the historical pick-up points belong to the historical delivery paths, that is, the historical pick-up points are the order areas within the historical delivery paths.

[0092] During the training of the path model, order areas within delivery regions need to be considered. Once the historical delivery routes and historical pick-up points of a delivery region are determined, there is no need to consider the order areas within that delivery region to ensure the effectiveness of generating historical delivery routes. Specifically, order areas within delivery regions where historical delivery routes have already been generated can be masked. Then, the path model outputs the historical delivery route of the next delivery end and the historical pick-up points of the corresponding delivery region for the next delivery end.

[0093] exist Figure 4 In this embodiment, the order area in the delivery area where a historical delivery route has been generated is hidden to ensure the validity of the historical delivery route output by the route model.

[0094] See Figure 5 That is, 500. Figure 5 This is a schematic diagram of the training path model according to an embodiment of the present invention. Figure 5 The training path model involves three time points: T=1, T=2, and T=3. At T=1, the connecting terminal picks up items from the order at historical connecting point 1; at T=2, the connecting terminal picks up items from the order at historical connecting point 2; and at T=3, the connecting terminal picks up items from the order at historical connecting point 3.

[0095] in, Figure 5 The Mask is used as a mask to block out the order areas in the delivery areas of the generated historical delivery routes, thus ensuring the validity of the historical delivery routes output by the path model. For example, during the training of the path model at T=2, the Mask is used to block out the order areas in the historical delivery routes output by the path model at T=1.

[0096] In one embodiment of the present invention, the order clustering region, order processing time and order region are input into the path model. After the path model is trained by the measurement gradient algorithm, the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region.

[0097] The REINFORCE algorithm, also known as the measurement gradient algorithm, optimizes route models to further improve delivery efficiency. REINFORCE is a policy gradient algorithm that aims to reduce the average delivery time of expected connecting orders by updating parameters through gradient ascent.

[0098] Specifically, the REINFORCE algorithm trains and optimizes the path model by iteratively sampling and evaluating delivery routes. In each iteration, a batch of delivery routes is generated, and the delivery time for orders on each route is calculated. Then, the gradient of each delivery route is calculated based on its performance, and the parameters of the path model are updated using gradient ascent. In this way, the path model is progressively optimized, enabling it to better select delivery routes.

[0099] S103. Input the order region and order processing time of the current order into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The current connection route constructed using the current delivery route and the current connection point is used to deliver the items of the current order.

[0100] After the path model is trained, the order region and order processing time of the current order can be input into the path model. The path model outputs the current delivery route and the current connection point of the delivery region, thereby enabling the delivery end to deliver the items of the current order.

[0101] See Figure 6 That is, 600. Figure 6 This is a schematic diagram illustrating the process of delivering items for the current order according to an embodiment of the present invention. Specifically, it includes the following steps:

[0102] S601. The delivery terminal retrieves the items for the current order at the current pick-up point according to the pick-up time.

[0103] The receiving end transports the items for the current order to the current receiving point according to the receiving time. The delivery end, at the current receiving point, can retrieve the items for the current order according to the receiving time.

[0104] The connection point moves from the current connection point to the next connection point according to the connection time of the next connection point.

[0105] S602. The delivery end delivers the items of the current order according to the current delivery route to meet the order processing time.

[0106] Once the delivery system obtains the items for the current order, it can deliver the items according to the current delivery route to meet the order processing time.

[0107] exist Figure 6 In one embodiment, the connecting end and the delivery end exchange the items in the order at the connecting point, and then the delivery end delivers the items in the order according to the current delivery route.

[0108] In the above embodiments of the present invention, historical orders are aggregated into multiple order clusters based on order region and order processing time. The order clusters, order processing time, and order regions are input into a path model. The path model outputs the historical delivery routes and historical connection points of the delivery regions, and the historical delivery routes and historical connection points satisfy the order processing time of the historical orders. The order region and order processing time of the current order are input into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point. Using a path model allows for direct knowledge of delivery routes and connection routes, thus accelerating the delivery speed and ensuring timely delivery.

[0109] See Figure 7 , Figure 7 This is a schematic diagram of the main structure of a device for delivering items according to an embodiment of the present invention. The device for delivering items can implement a method for delivering items, such as... Figure 7 As shown in Figure 700, the device for delivering items specifically includes:

[0110] The aggregation module 701 is used to aggregate historical orders into multiple order clustering regions based on the order region and the order processing time.

[0111] Training module 702 is used to input the order clustering region, the order processing time and the order region into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order.

[0112] The output module 703 is used to input the order region and order processing time of the current order into the path model. The path model outputs the current delivery route and the current connection point of the delivery region, and delivers the items of the current order using the current connection route constructed by the current delivery route and the current connection point.

[0113] The aggregation module 701 is specifically used to divide the historical orders into orders in multiple time periods based on the order processing time; and to aggregate orders in the same time period based on the order region to obtain the multiple order cluster regions.

[0114] In one embodiment of the present invention, the training module 702 is specifically used to input the order clustering region, the order processing time and the order region into the path model, and the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region;

[0115] The historical pick-up points in the delivery area meet the pick-up times of the historical orders;

[0116] The actual processing time of the historical orders is determined in the historical delivery route to meet the order processing time requirements of the historical orders.

[0117] The training module 702 is specifically used to ensure that the historical pick-up points in the delivery area meet the pick-up times of the historical orders in multiple historical delivery routes.

[0118] In one embodiment of the present invention, the training module 702 is specifically used to input the order clustering region, the order processing time, and the order region into the path model;

[0119] The path model outputs the historical delivery path of a delivery terminal and the historical connection point of the corresponding delivery area. After hiding the order area in the delivery area, the path model outputs the historical delivery path of the next delivery terminal and the historical connection point of the corresponding delivery area of ​​the next delivery terminal.

[0120] In one embodiment of the present invention, the training module 702 is specifically used to input the order clustering region, the order processing time and the order region into the path model, and after training the path model with a measurement gradient algorithm, the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region.

[0121] In one embodiment of the present invention, the output module 703 is specifically used for the delivery terminal to obtain the items of the current order at the current pick-up point according to the pick-up time;

[0122] The delivery terminal delivers the items of the current order according to the current delivery route to meet the order processing time.

[0123] Figure 8 An exemplary system architecture 800 is shown, which can be applied to a method or apparatus for delivering items according to embodiments of the present invention.

[0124] like Figure 8 As shown, system architecture 800 may include terminal devices 801, 802, and 803, a network 804, and a server 805. Network 804 serves as the medium for providing communication links between terminal devices 801, 802, and 803 and server 805. Network 804 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0125] Users can use terminal devices 801, 802, and 803 to interact with server 805 via network 804 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 801, 802, and 803, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0126] Terminal devices 801, 802, and 803 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0127] Server 805 can be a server providing various services, such as a backend management server supporting shopping websites browsed by users using terminal devices 801, 802, and 803 (for example only). The backend management server can analyze and process data such as received product information query requests, and feed back the processing results (such as target push information and product information—for example only) to the terminal devices.

[0128] It should be noted that the method for delivering items provided in this embodiment of the invention is generally executed by server 805, and correspondingly, the device for delivering items is generally located in server 805.

[0129] It should be understood that Figure 8 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0130] The following is for reference. Figure 9 It shows a schematic diagram of the structure of a computer system 900 suitable for implementing a terminal device of the present invention. Figure 9 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0131] like Figure 9 As shown, the computer system 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 902 or programs loaded from storage section 908 into random access memory (RAM) 903. The RAM 903 also stores various programs and data required for the operation of the system 900. The CPU 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0132] The following components are connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 910 as needed so that computer programs read from it can be installed into storage section 908 as needed.

[0133] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 909, and / or installed from removable medium 911. When the computer program is executed by central processing unit (CPU) 901, it performs the functions defined above in the system of this invention.

[0134] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0135] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0136] The modules described in the embodiments of the present invention can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor can be described as including an aggregation module, a training module, and an output module. The names of these modules do not necessarily limit the module itself; for example, the aggregation module can also be described as "used to aggregate historical orders into multiple order clustering regions based on order region and order processing time".

[0137] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include:

[0138] Based on the order region and order processing time, historical orders are aggregated into multiple order clusters.

[0139] The order clustering region, the order processing time, and the order region are input into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order.

[0140] The order region and order processing time of the current order are input into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point.

[0141] According to the technical solution of this invention, historical orders are aggregated into multiple order clustering regions based on order region and order processing time. The order clustering regions, order processing time, and order regions are input into a path model. The path model outputs the historical delivery route of the delivery end and the historical connection points of the delivery region, and the historical delivery route and the historical connection points satisfy the order processing time of the historical orders. The order region and order processing time of the current order are input into the path model. The path model outputs the current delivery route of the delivery end and the current connection point of the delivery region. The items of the current order are delivered using the current connection path constructed from the current delivery route and the current connection point. Using a path model allows for direct knowledge of the delivery route and connection path, thus accelerating the delivery speed and ensuring timely delivery.

[0142] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention. It should be noted that the acquisition, storage, and application of user personal information involved in the technical solutions of this disclosure comply with relevant laws and regulations and do not violate public order and good morals.

Claims

1. A method for delivering goods, characterized in that, include: Based on the order region and order processing time, historical orders are aggregated into multiple order clusters. The order clustering region, the order processing time, and the order region are input into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order. The path model includes an encoder and a decoder. The encoder is based on an attention mechanism, and the decoder is based on a multi-head attention mechanism and a single-head attention mechanism. The training of the path model requires the use of a gradient measurement algorithm and a masking mechanism. The order region and order processing time of the current order are input into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point.

2. The method for delivering items according to claim 1, characterized in that, Based on order region and order processing time, historical orders are aggregated into multiple order cluster regions, including: Based on the order processing time, the historical orders are divided into orders in multiple time periods; Based on the order region, orders within the same time period are aggregated to obtain the multiple order cluster regions.

3. The method for delivering items according to claim 1, characterized in that, The step of inputting the order clustering region, the order processing time, and the order region into the path model, wherein the path model outputs the historical delivery route of the delivery end and the historical pick-up point of the delivery region, and the historical delivery route and the historical pick-up point satisfy the order processing time of the historical order, includes: The order clustering region, the order processing time, and the order region are input into the path model, and the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region. The historical pick-up points in the delivery area meet the pick-up times of the historical orders; The actual processing time of the historical orders is determined in the historical delivery route to meet the order processing time requirements of the historical orders.

4. The method for delivering items according to claim 3, characterized in that, The historical pick-up points in the delivery area that meet the pick-up times of the historical orders include: The historical pick-up points in the delivery area meet the pick-up times of the historical orders in multiple historical delivery routes.

5. The method for delivering items according to claim 1, characterized in that, The step of inputting the order clustering region, the order processing time, and the order region into the path model, and the path model outputting the historical delivery route of the delivery end and the historical connection point of the delivery region, includes: Input the order clustering region, the order processing time, and the order region into the path model; The path model outputs the historical delivery path of a delivery terminal and the historical connection point of the corresponding delivery area. After hiding the order area in the delivery area, the path model outputs the historical delivery path of the next delivery terminal and the historical connection point of the corresponding delivery area of ​​the next delivery terminal.

6. The method for delivering items according to claim 1, characterized in that, The step of inputting the order clustering region, the order processing time, and the order region into the path model, and the path model outputting the historical delivery route of the delivery end and the historical connection point of the delivery region, includes: The order clustering region, the order processing time, and the order region are input into the path model. After training the path model with a gradient measurement algorithm, the path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region.

7. The method for delivering items according to claim 1, characterized in that, The current delivery route and the current connection point together form the current connection route for delivering the items of the current order, including: The delivery terminal retrieves the items for the current order at the current pick-up point according to the pick-up time; The delivery terminal delivers the items of the current order according to the current delivery route to meet the order processing time.

8. A device for delivering goods, characterized in that, include: The aggregation module is used to aggregate historical orders into multiple order clusters based on order region and order processing time; The training module is used to input the order clustering region, the order processing time, and the order region into the path model. The path model outputs the historical delivery route of the delivery end and the historical connection point of the delivery region, and the historical delivery route and the historical connection point satisfy the order processing time of the historical order. The path model includes an encoder and a decoder. The encoder is based on an attention mechanism, and the decoder is based on a multi-head attention mechanism and a single-head attention mechanism. The training of the path model requires the use of a gradient measurement algorithm and a masking mechanism. The output module is used to input the order region and order processing time of the current order into the path model. The path model outputs the current delivery route and the current connection point of the delivery region. The items of the current order are delivered using the current connection route constructed from the current delivery route and the current connection point.

9. An electronic device for delivering goods, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-7.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images