Method and platform for dynamically combining e-commerce order and same-city order delivery paths

Abstract

The application relates to an e-commerce order and same-city order dynamic joint distribution path optimization method and platform. The optimization method comprises the following steps: when a distribution request of a newly added same-city order distribution is received, acquiring first joint distribution information currently executed; from the first joint distribution information, acquiring all first distribution paths currently, and using RIM of an ALNS-RIM hybrid optimization algorithm to detect the first distribution path capable of inserting the newly added same-city order to obtain a candidate distribution path, and detecting a candidate path segment corresponding to the insertion of the newly added same-city order in the candidate distribution path; based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segment, performing insertion path planning, generating a target path segment, and updating all first distribution paths according to the target path segment to generate a path optimization result comprising second joint distribution information. Through the application, the problem that the scheme for jointly distributing e-commerce orders and same-city orders in related technologies is not timely and has low distribution efficiency is solved.

Description

Technical Field

[0001] This application relates to the field of computer intelligent application technology, and in particular to a method and platform for optimizing the dynamic joint delivery route of e-commerce orders and same-city orders. Background Technology

[0002] In last-mile delivery, to improve delivery efficiency, it is often considered to combine bulk e-commerce orders with scattered local orders for joint delivery. In related technologies, for local orders that do not require immediate delivery, e-commerce orders and local orders can be planned and coordinated together in advance. However, for local orders that require immediate pickup and delivery, it is necessary to consider how to incorporate new local orders into the already planned scheme. Two methods are used in related technologies: one is a delayed processing strategy, where all new orders are delayed and treated as static. However, this strategy reduces resource utilization and delivery efficiency. The other is to divide the time period into several equal time intervals, re-planning new and undelivered orders at each time interval. However, the rolling time domain method cannot respond to customer needs in a timely manner, leading to unfinished delivery tasks. Furthermore, when handling large-scale order deliveries, the re-planning results cannot promptly guide actual scheduling, resulting in poor practicality.

[0003] Currently, solutions for joint delivery of e-commerce and local orders using related technologies suffer from issues such as untimely response and low delivery efficiency, and no effective solutions have yet been proposed. Summary of the Invention

[0004] This application provides a method and platform for optimizing the dynamic joint delivery route of e-commerce orders and local orders, so as to at least solve the problems of untimely response and low delivery efficiency in the joint delivery schemes of e-commerce orders and local orders in related technologies.

[0005] In a first aspect, embodiments of this application provide a method for optimizing the dynamic joint delivery route of e-commerce orders and same-city orders, comprising: upon receiving a delivery request for a newly added same-city order, obtaining currently executed first joint delivery information, wherein the first joint delivery information is generated by using an ALNS-RIM hybrid optimization algorithm composed of a real-time insertion method and an adaptive large-domain search algorithm to dynamically optimize the joint delivery route of historical joint delivery information and the previous newly added same-city order; obtaining all current first delivery routes from the first joint delivery information, and using the RIM of the ALNS-RIM hybrid optimization algorithm to detect the first delivery routes that can be inserted into the newly added same-city order to obtain candidate delivery routes, and detecting candidate path segments corresponding to the insertion of the newly added same-city order in the candidate delivery routes; performing insertion path planning based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments to generate target path segments, and updating all first delivery routes according to the target path segments to generate a path optimization result including second joint delivery information, wherein the second joint delivery information includes a target delivery route corresponding to the target delivery route segment and the unupdated first delivery routes.

[0006] Secondly, embodiments of this application provide a service platform, including a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to perform the steps of the dynamic joint delivery route optimization method for e-commerce orders and same-city orders described in the first aspect.

[0007] Compared to related technologies, the e-commerce order and same-city order dynamic joint delivery route optimization method and platform provided in this application embodiment adopts the following approach: upon receiving a delivery request for a newly added same-city order, the first joint delivery information currently being executed is obtained. This first joint delivery information is generated by using the ALNS-RIM hybrid optimization algorithm, which combines real-time insertion and adaptive large-domain search, to dynamically optimize the joint delivery route of historical joint delivery information and the previous newly added same-city order. From the first joint delivery information, all current first delivery routes are obtained, and the RIM of the ALNS-RIM hybrid optimization algorithm is used to detect the first delivery routes that can be inserted into the newly added same-city order, thereby obtaining candidate delivery routes. In the candidate delivery routes, candidate route segments corresponding to the newly added same-city orders are detected; based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate route segments, insertion route planning is performed to generate target route segments, and all first delivery routes are updated according to the target route segments to generate route optimization results including second joint delivery information. The new same-city orders are inserted using a real-time insertion method, and the optimal delivery route for the inserted new same-city orders is planned using the ALNS-RIM hybrid optimization algorithm. This solves the problems of untimely response and low delivery efficiency in the joint delivery scheme of e-commerce orders and same-city orders in related technologies, and achieves the beneficial effects of reducing delivery costs, improving delivery vehicle utilization and delivery efficiency.

[0008] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0010] Figure 1 This is a hardware structure block diagram of the terminal of the dynamic joint delivery route optimization method for e-commerce orders and same-city orders according to an embodiment of this application;

[0011] Figure 2 This is a flowchart of a method for optimizing the dynamic joint delivery route of e-commerce orders and same-city orders according to an embodiment of this application;

[0012] Figure 3 This is a schematic diagram showing the insertion of a new local order in an embodiment of this application;

[0013] Figure 4 This is a structural block diagram of the device for dynamic joint delivery route optimization of e-commerce orders and same-city orders according to an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application. Furthermore, it is understood that although the efforts made in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, modifications to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0015] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0016] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "a," "an," "an," "the," and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms "comprising," "including," "having," and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. "Multiple stages" used in this application refers to two or more stages. "And / or" describes the relationship between related objects, indicating that three relationships may exist; for example, "A and / or B" can represent: A alone, A and B simultaneously, and B alone. The terms "first," "second," "third," etc., used in this application are merely to distinguish similar objects and do not represent a specific ordering of objects.

[0017] Before describing the specific embodiments of this application, the difficulties overcome in proposing the method for dynamic joint delivery route optimization of e-commerce orders and same-city orders in this application are explained as follows:

[0018] For same-city orders requiring real-time pickup and delivery, compared to the initial static joint delivery problem (which involves planning e-commerce orders and same-city orders together in advance), inserting new same-city orders requires effectively handling these new orders added during the delivery process, as they were not included in the initial statically planned path. To ensure efficient completion of delivery tasks, new same-city orders can be integrated into the initial static joint delivery path, the path can be updated, and the delivery can be completed. This allows for the development of a path delivery strategy that satisfies the pickup and delivery needs and time window constraints of all orders, thereby optimizing the overall objective function. Therefore, how to reasonably insert newly added same-city orders into the initial path requires overcoming the following two difficulties: First, determining the timing for inserting new same-city orders is difficult. For newly added same-city orders, it is necessary to obtain information about the new same-city orders in real time and determine the appropriate time to insert them. Adding new same-city orders inherently has strict time windows, and inserting new orders inevitably affects the delivery distance, delivery time, and delivery cost of the initial delivery route, as well as the delivery time for other customers. Therefore, choosing a suitable method to determine the timing of inserting new same-city orders is crucial. Secondly, determining the insertion location for new same-city orders is difficult. Since same-city orders require pickup before delivery by the same vehicle, it's necessary to simultaneously insert the pickup and delivery points of the same-city order into the initial route, while also achieving the optimal overall result for both customer points. The location of the new same-city order might deviate significantly from the initial delivery route, making it unsuitable to insert it into the initial delivery route; in this case, a new vehicle should be dispatched for delivery. Therefore, determining the insertion location is another challenge for inserting new same-city orders.

[0019] The method embodiments provided in this example can be executed on a terminal, computer, or similar computing device. Taking running on a terminal as an example, Figure 1 This is a hardware structure block diagram of the terminal for the dynamic joint delivery route optimization method for e-commerce orders and same-city orders according to an embodiment of this application. For example... Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. Optionally, the terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0020] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the dynamic joint delivery route optimization method for e-commerce orders and same-city orders in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0021] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0022] This embodiment provides a method for optimizing the dynamic joint delivery route of e-commerce orders and same-city orders running on the aforementioned terminal. Figure 2 This is a flowchart of the dynamic joint delivery route optimization method for e-commerce orders and same-city orders according to an embodiment of this application, such as... Figure 2 As shown, the process includes the following steps:

[0023] Step S201: When a delivery request for a newly added same-city order is received, the first joint delivery information currently being executed is obtained. The first joint delivery information is generated by using the ALNS-RIM hybrid optimization algorithm, which is composed of real-time insertion method and adaptive large-domain search algorithm, to dynamically optimize the joint delivery path of historical joint delivery information and the previous newly added same-city order.

[0024] In this embodiment, when performing the corresponding dynamic joint delivery optimization, it is first necessary to process the order information of the received orders (including e-commerce orders and local orders during the initial planning, as well as newly added local orders in real time) to determine the location of the delivery node corresponding to the order, the service time window, and the distance between the corresponding delivery nodes. In this embodiment, the corresponding data processing includes calculating the latitude and longitude distance based on the latitude and longitude parameters (latitude and longitude coordinates) in the order information and converting the demand service time in the order information into a service time window. At the same time, the quantity of goods to be picked up and delivered corresponding to the order is also obtained from the order information. Of course, the target information of the available delivery vehicles (including the number of vehicles, the maximum load capacity of the vehicles, the unit usage cost, and the unit driving cost) is also obtained from the preset delivery resources. After obtaining the relevant order information and delivery resource information, the corresponding dynamic joint delivery planning is performed, thereby generating the corresponding joint delivery information.

[0025] In this embodiment, during the joint delivery of e-commerce orders and same-city orders according to all the first delivery routes corresponding to the planned first joint delivery information, the occurrence of new same-city orders is monitored in real time. If no new same-city orders occur, the delivery plan corresponding to the first joint delivery information is executed until the delivery task is completed. When a new same-city order occurs, dynamic joint delivery optimization of e-commerce orders and same-city orders is initiated, that is, the corresponding new same-city order is inserted into the currently executed first delivery route. Therefore, it is necessary to obtain the currently executed first joint delivery information.

[0026] Step S202: Obtain all current first delivery routes from the first joint delivery information, and use the RIM of the ALNS-RIM hybrid optimization algorithm to detect first delivery routes into which new same-city orders can be inserted, thereby obtaining candidate delivery routes, and detecting candidate route segments corresponding to the insertion of new same-city orders in the candidate delivery routes.

[0027] In this embodiment, when a new same-city order occurs, it is necessary to reasonably insert a set of delivery nodes (including pickup and delivery nodes) corresponding to the new same-city order into the same first delivery path. To achieve fast and accurate insertion and improve delivery decision efficiency, a feasible candidate delivery path for insertion is determined from all first delivery paths based on the delivery location corresponding to the new same-city order. In this embodiment, the Real-time Insertion Method (ALNS-RIM) of the hybrid optimization algorithm is used. (abbreviated as RIM) From all first delivery routes, corresponding candidate delivery routes are selected. At the same time, after selecting candidate delivery routes, while adhering to the constraints of service time window and vehicle capacity, candidate path segments that can insert a set of delivery nodes corresponding to the new same-city order are selected from each candidate delivery route. In this embodiment, considering the strict time requirements of new same-city orders, delivery path segments that have already been served (that is, inserted between delivery nodes to be delivered) are excluded when inserting them into the path, and delivery path segments that exceed the new service time window are not considered. The path formed by the delivery nodes covered within the new service time window is the candidate path segment. It should be noted that the location of the delivery node corresponding to the order is fixed. Therefore, the location of the delivery node insertion described in this embodiment does not refer to the actual spatial location, but to the order of pickup and delivery when carrying out logistics delivery.

[0028] Step S203: Based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, perform insertion path planning to generate target path segments, and update all first delivery paths according to the target path segments to generate path optimization results including second joint delivery information. The second joint delivery information includes the target delivery path corresponding to the target delivery segment and the unupdated first delivery path.

[0029] In this embodiment, after determining the candidate path segments that can be inserted, at least one path planning method corresponding to the ALNS-RIM hybrid optimization algorithm is adopted to select the position of a set of delivery nodes corresponding to the new same-city order on the candidate path segment. With the goal of minimizing the change cost caused by the insertion of the order, the optimal insertion position is determined and the corresponding optimal path segment, that is, the target path segment, is generated. Then, the target path segment is used as the vehicle route for delivering the new same-city order and the original order.

[0030] In this embodiment, the unupdated first delivery path and the target delivery path are used as the second joint delivery information obtained by completing the current optimization. In this embodiment, for the first delivery path that is not selected as a candidate delivery path and for which no candidate path segment is selected, the default is to continue order delivery according to the original first delivery path. Of course, the first delivery path can also be optimized, but this will result in ineffective computing power or processing resources being wasted.

[0031] In this embodiment, after generating the target path segment, the original first delivery path associated with the target path segment is updated. That is, orders to be delivered after the current time will be delivered according to the target path segment. However, the corresponding first delivery path also includes a part of the path segment that has been delivered. By updating the corresponding first delivery path, a new delivery path, namely the target delivery path, is generated. Then, based on the target path segment of the target delivery path, the existing orders to be delivered and the new same-city orders are delivered.

[0032] Through steps S201 to S203, upon receiving a delivery request for a new same-city order, the system obtains the currently executing first joint delivery information; from the first joint delivery information, it obtains all current first delivery routes and uses the RIM of the ALNS-RIM hybrid optimization algorithm to detect first delivery routes into which the new same-city order can be inserted, obtaining candidate delivery routes; and from the candidate delivery routes, it detects candidate path segments corresponding to the inserted new same-city order; based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, it performs insertion path planning, generates target path segments, and updates all first delivery routes according to the target path segments, generating path optimization results including second joint delivery information. By introducing a real-time insertion method to insert new same-city orders and using the ALNS-RIM hybrid optimization algorithm to plan the optimal delivery route for inserting new same-city orders, the system solves the problems of untimely response and low delivery efficiency in related technologies for joint delivery of e-commerce orders and same-city orders, achieving the beneficial effects of reducing delivery costs, improving delivery vehicle utilization, and increasing delivery efficiency.

[0033] It should be noted that, in this embodiment, to address the difficulty of determining the timing for inserting new same-city orders, a Real-Time Injection (RIM) algorithm is introduced, and an ALNS-RIM hybrid optimization algorithm is proposed for planning. To address the difficulty of determining the location for inserting new same-city orders, an insertion rule is established based on real-time insertion, thereby reasonably inserting new same-city orders without changing the initial delivery route. The insertion steps are as follows: Step 1: Determine candidate delivery routes that can be inserted based on the location of the delivery node corresponding to the new same-city order; Step 2: Determine alternative route segments that can be inserted based on the service time window limit of the delivery node corresponding to the new same-city order; Step 3: Determine the insertion cost corresponding to all alternative route segments that can be inserted, and select the location with the lowest insertion cost for insertion; Step 4: If no location can be inserted, a new... Delivery vehicles make deliveries. In this embodiment, during the process of determining insertable candidate delivery routes, a reasonable candidate list is set, and the distance from the target central node (delivery center) to the new order is considered as the maximum reasonable distance. A circle is drawn with the new order as the center and the distance as the radius. The route covered by the semicircle is the candidate route. In this embodiment, considering the strict timeliness requirements of new same-city orders, delivery route segments that have already been served are excluded when inserting the route, and delivery route segments that exceed the new service time window corresponding to the new delivery node are also not considered. The route formed by the delivery nodes covered within the new time window corresponding to the new same-city order is the candidate route segment. This processing method helps to further reduce the number of delivery nodes on the candidate route, thereby improving the calculation speed and processing orders in a timely manner.

[0034] It should be further explained that the dynamic joint delivery route optimization method for e-commerce orders and local orders in this application embodiment also produces the following beneficial effects: First, this application embodiment adopts a globally unified delivery method that integrates bulk e-commerce orders and scattered local orders at the last mile. By establishing a suitable mathematical model and using the ALNS-RIM hybrid optimization algorithm to plan the optimal delivery route, it can effectively improve vehicle utilization, integrate logistics resources, and thus improve the operational capabilities of the corresponding logistics delivery operators. Second, this application embodiment considers the joint delivery route optimization problem of scattered local orders when considering the last-mile delivery of long-distance, large-volume e-commerce express orders, providing a reference for relevant operators involved in cross-border and long-distance transportation and last-mile delivery, so as to encourage them to integrate logistics resources and reduce delivery costs.

[0035] Figure 3 This is a schematic diagram illustrating the insertion of a new local order in an embodiment of this application. (Refer to...) Figure 3 In some alternative implementations, a set of delivery nodes (see reference) is used to include new same-city orders. Figure 3 Pickup node 3 + and receiving node 3 -Simultaneously, it inserts the order into the first delivery route (corresponding to the initial delivery route), adhering to pickup and delivery pairing, service time window, and vehicle capacity constraints. For example, when a new same-city order (corresponding to a real-time same-city order) appears at a certain time, and the k-th delivery vehicle is traveling on the planned route (the first delivery route), it checks all positions in the first delivery route where the new same-city order can be inserted and selects the position with the lowest insertion cost to perform the insertion operation. For the specific insertion process, refer to [reference needed]. Figure 3 Assume the first delivery route of the kth delivery vehicle is 1. + →2 + →1 - →N + →2 - →N - At time t, a new same-city order 3 is added (the corresponding delivery node includes pickup node 3). + and receiving node 3 - ), perform insertion path planning, and generate the target delivery path as: 1 + →2 + →1 - →N + →2 - →3 + →N - →3 - .

[0036] In some embodiments, the RIM of the ALNS-RIM hybrid optimization algorithm is used to detect the first delivery route for inserting new same-city orders and obtain candidate delivery routes, which is achieved through the following steps:

[0037] Step 21: Obtain the order information for each new same-city order from the delivery request. The order information includes the new pickup and delivery node pair corresponding to the new same-city order. The new pickup and delivery node pair is used to represent the pickup node and delivery node corresponding to the new same-city order.

[0038] In this embodiment, the new same-city order is a one-to-one pickup and delivery relationship, that is, one new same-city order corresponds to one pickup node and one corresponding delivery node. In this embodiment, by performing corresponding data processing on the order information, the location (latitude and longitude parameters) of the pickup node and delivery node corresponding to the new same-city order, the pickup volume of the pickup node and the new service time window (pickup time period), and the delivery volume of the delivery node and the new service time window (delivery time period) can be determined.

[0039] Step 22: Determine the pickup node and receiving node for each new same-city order corresponding to the new pickup and delivery node pair, and determine the distance between the pickup node and receiving node corresponding to the same new same-city order and the target center node, to obtain the first path distance and the second path distance.

[0040] Step 23: Take the longest path distance between the first path distance and the second path distance as the target distance, and after determining the target node corresponding to the target distance, determine the preset area based on the target node and the target distance, wherein the center of the preset area is located at the target node position.

[0041] Step 24: Select the first delivery route within the preset area from all first delivery routes to obtain candidate delivery routes.

[0042] In this embodiment, the distances between a group of delivery nodes corresponding to a new same-city order and the target central node (corresponding to the delivery center) are considered to divide the corresponding area, thereby determining the first delivery path within the divided area as a candidate path. In this embodiment, the maximum value among the distances between the target central node and the group of delivery nodes corresponding to the new same-city order is used as the target distance, and a semicircle is drawn with the delivery node corresponding to the target distance as the center and the target distance as the radius. The drawn semicircle is the corresponding preset area, and the first delivery path covered by the preset area is used as the optional candidate delivery path.

[0043] Through steps 21 to 24 above, the location of a set of delivery nodes corresponding to the new same-city order is used to determine the preliminary candidate delivery path that can be inserted, thereby reducing the number of candidate locations that can be inserted on the candidate path, thus improving the calculation speed and processing the order in a timely manner.

[0044] In some embodiments, the order information also includes the new service time windows corresponding to the pickup node and the receiving node. In the candidate delivery routes, the candidate route segment corresponding to the newly added same-city order is detected and inserted, which is achieved through the following steps:

[0045] Step 31: Obtain the new service time window of the pickup node and the receiving node corresponding to the new same-city order from the order information, detect the undelivered first delivery node from all the first delivery nodes corresponding to each candidate delivery path, and determine the path segment where all the undelivered first delivery nodes are located from the candidate delivery paths to obtain the candidate path segment.

[0046] Step 32: After determining the service time window corresponding to each first delivery node on the candidate route segment, select alternative route segments from all candidate route segments based on the time difference between the corresponding new service time window and each service time window.

[0047] In some alternative implementations, candidate path segments are selected from all candidate path segments based on the new service time window and the time difference of each service time window, through the following steps:

[0048] Step 321: Based on the time difference between each corresponding service time window and the newly added service time window, determine whether all service time windows corresponding to the candidate path segment exceed the newly added service time window.

[0049] Step 322: If it is determined that none of the service time windows corresponding to the candidate path segment have exceeded the newly added service time window, the candidate path segment is determined as the alternative path segment.

[0050] Step 33: Based on the total pickup and delivery demand corresponding to the alternative route segments and the current cargo capacity of the corresponding delivery vehicles, determine the cargo capacity of the corresponding delivery vehicles. The current cargo capacity is used to represent the remaining cargo capacity of the delivery vehicles after loading the goods of the first delivery node that has not yet been delivered.

[0051] Step 34: Among the alternative route segments, select alternative route segments with a carrying capacity not less than the new pickup and delivery demand obtained from the order information to obtain candidate route segments. The new pickup and delivery demand is used to represent the amount of goods collected from the pickup node or delivered to the receiving node.

[0052] In this embodiment, considering the strict timeliness requirements of new same-city orders, delivery route segments that have already been served are excluded when inserting them into the path, and delivery route segments that exceed the new service time window corresponding to the new delivery node are also not considered. The path formed by the delivery nodes covered within the new time window corresponding to the new same-city order is the alternative path segment.

[0053] Through steps 31 to 34 above, the process of selecting a preliminary route segment from the candidate delivery routes to insert new same-city orders is realized, reducing the number of alternative locations for a set of delivery nodes that allow new same-city orders to be inserted, and improving the efficiency of optimization decision-making.

[0054] In some embodiments, the following steps are performed before obtaining candidate path segments:

[0055] Step 41: If the available capacity is less than the new pickup and delivery demand or if at least one service time window exceeds the new service time window, select a new delivery vehicle based on the new pickup and delivery demand corresponding to the new same-city order.

[0056] Step 42: Using at least the pickup and delivery nodes corresponding to the new same-city orders as the corresponding delivery nodes, generate a new delivery route corresponding to the new same-city orders using the ALNS-RIM hybrid optimization algorithm.

[0057] Step 43: Generate the second joint delivery information by combining the new delivery route and all first delivery routes.

[0058] In this embodiment, if a new same-city order cannot be inserted into an existing delivery route due to vehicle capacity and time window constraints, a new delivery vehicle needs to be added to serve the order. If there is no optimal insertion location, a new vehicle is added to serve the order, forming a new delivery route. In this embodiment, when planning a new delivery route, ALNS is used for large neighborhood search and destruction repair operations. At the same time, multiple new same-city orders are considered for unified planning to make full use of delivery vehicle resources and improve delivery vehicle utilization.

[0059] In some embodiments, all first delivery routes are updated based on the target path segment by the following steps:

[0060] Step 51: Among all the first delivery routes, detect the first delivery route associated with the target delivery route to obtain the delivery route to be updated.

[0061] Step 52: Determine the candidate path segments for each delivery route to be updated, and update the corresponding candidate path segments to the target path segments to obtain the target delivery route.

[0062] Step 53: Generate the second joint delivery information by combining the target delivery route and the unupdated first delivery route.

[0063] Through steps 51 to 53 above, the target delivery route and the second joint delivery information are generated based on the target route segment and the first delivery route, and the dynamic joint delivery route optimization results are output.

[0064] In some embodiments, based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, insertion path planning is performed to generate the target path segment, including the following steps:

[0065] Step 61: Determine all first delivery nodes corresponding to the candidate path segments, and determine the pickup node and receiving node corresponding to at least one new same-city order associated with each candidate path segment, to obtain a delivery node set, wherein the delivery nodes in the delivery node set include all first delivery nodes, all pickup nodes, and all receiving nodes.

[0066] Step 62: Process the delivery node set according to the preset path planning method to generate candidate path data. The path planning method includes one of the following: using ALNS of the ALNS-RIM hybrid optimization algorithm to perform neighborhood search iteration on the delivery nodes in the delivery node set; or using RIM of the ALNS-RIM hybrid optimization algorithm to perform node insertion traversal.

[0067] In some alternative implementations, a path planning method using RIM with node insertion traversal based on the ALNS-RIM hybrid optimization algorithm is employed to process the delivery node set to generate candidate path data, including the following steps:

[0068] Step 621-1: Using the RIM of the ALNS-RIM hybrid optimization algorithm, determine the two adjacent first delivery nodes located on the candidate path segment, and set the interval between the two adjacent first delivery nodes as the interval to be inserted.

[0069] Step 621-2: Randomly insert the pickup node and receiving node corresponding to the same newly added same-city order into at least one insertion interval, and generate a first initial path, wherein the pickup node is inserted before the receiving node.

[0070] Step 621-3: Repeat the process of inserting at least one of the pickup node and the receiving node in all the intervals to be inserted to obtain multiple first initial paths, and use these multiple first initial paths as candidate path data.

[0071] In some alternative implementations, an ALNS-RIM hybrid optimization algorithm is used to perform neighborhood search iterations on the delivery node set to process the delivery node set and generate candidate path data, including the following steps:

[0072] Step 622-1: According to the preset construction method, construct paths for all delivery nodes in the delivery node set to generate multiple second initial paths. In the second initial paths, the pickup node is located before the receiving node. The construction method includes one of the following: random insertion method, nearest insertion method, and greedy insertion method.

[0073] In this embodiment, a delivery node from the delivery node set is used as the solution for path construction initialization. This means that a delivery node (pickup node or delivery node) corresponding to the order is inserted into a second initial path according to a predefined method.

[0074] Step 622-2: Using the roulette wheel algorithm, select the target destruction operation and the target repair operation from the various destruction operations and various repair operations corresponding to ALNS in the ALNS-RIM hybrid optimization algorithm. The destruction operation represents the removal of the delivery node in the second initial path. The destruction operation includes one of the following: random removal, worst removal, and similar removal. The repair operation represents the insertion of the removed delivery node into the second initial path after the removal is completed. The repair operation includes one of the following: random insertion and greedy insertion.

[0075] Step 622-3: Perform a target repair operation on the reconstructed path obtained by performing the target destroy operation on the second initial path to generate the third path.

[0076] In this embodiment, by dynamically adjusting the weights and selecting the "destroy" or "reconstruct" operators, ALNS assigns a weight to each destroy and repair operator. These weights control the frequency of use of each destroy and repair operator during the search process. During the search process, ALNS dynamically adjusts the weights of various destroy and repair operators according to various rules in order to obtain a better neighborhood solution. The two key aspects include the selection of each operator and the continuous updating of the operator weights.

[0077] In this embodiment, there are multiple choices in both the destroy and repair operators, and each operator is assigned a specific weight. Based on these operator weights, a roulette wheel selection method is used to select the destroy and repair operators, and the number of times the operators are used is updated. The roulette wheel selection steps are as follows:

[0078] Step 1: Calculate the selection probability based on the operator weights: .

[0079] Step 2: Calculate the cumulative probability of each operator: .

[0080] Step 3: Perform the selection operation: Generate a pseudo-random number r that is uniformly distributed in the interval [0, 1]; if If the condition is met, then select individual 1; otherwise, select individual k such that: Established;

[0081] Step 4: Repeat the operation: The operation should be performed once for both the "destroy" and "reconstruct" operators.

[0082] In this embodiment, the methods for destroying the solution mainly include random removal, worst-case removal, and similarity removal. Random removal refers to deleting any delivery node in the current solution; worst-case removal is to delete the longest road segment in the current solution; similarity removal is to remove delivery nodes in the current solution based on the similarity between orders; the methods for reconstructing the solution include, but are not limited to, random insertion and greedy insertion. Random insertion means that the nodes to be removed are inserted one by one into any place in the destroyed solution; greedy insertion is to place the nodes to be removed in the place with the lowest distance cost, that is, the place with the shortest total path after insertion.

[0083] Step 622-4: Based on the objective function, determine the variable cost corresponding to the third path, and perform destroy and repair operations on the third path according to the variable cost until a fourth path with a variable cost less than a preset cost threshold is generated, thus obtaining candidate path data, wherein the candidate path data includes all fourth paths.

[0084] It should be noted that the neighborhood search iteration operation performed by ALNS in this application embodiment is clear and feasible to those skilled in the art, and the related operations adopted in this application embodiment do not constitute a limitation on the understanding of this application embodiment.

[0085] Step 63: Determine the variable cost corresponding to the planned path corresponding to the candidate path data according to the preset objective function, and select the planned path with the minimum variable cost from all the planned paths corresponding to the candidate path data to obtain the target path segment. The objective function is constructed based on the increased delivery cost of delivery vehicles and the late delivery penalty cost caused by violating the time window. The delivery distance corresponding to the increased delivery cost of delivery vehicles is determined based on the latitude and longitude of the corresponding delivery node.

[0086] In this embodiment, during the dynamic joint delivery route optimization process, the process of solving the optimal delivery route is mathematically modeled, and the following settings and processing are performed regarding the assumptions, parameters, objective function of the model, and constraints:

[0087] First, the following settings are made: When a new same-city order is generated, the target central node obtains information such as the customer's pickup and delivery demand, the geographical location of a set of delivery nodes corresponding to the new same-city order, and the new service time window for each delivery node through intelligent information technology; the new same-city order has high timeliness requirements and is set as a time window constraint; if the service time window limit is exceeded, it will not be inserted into the existing delivery route, but will be delivered by a new delivery vehicle according to the planned new delivery route; at the same time, the pickup node and delivery node of the new same-city order must be inserted into the same initial delivery route.

[0088] Secondly, set the following parameters: Add local orders , This indicates the number of newly added same-city orders; pickup node. Delivery point ; and Let these represent the sets of newly added pickup points and newly added delivery points, respectively. + = This represents the set of nodes corresponding to newly added same-city orders.

[0089] Next, we establish the objective function: Considering the objective function for joint delivery of new same-city orders, it consists of two parts: the total cost of the initial static planning delivery and the variable cost brought about by inserting all new same-city orders. The total cost is the sum of these two parts, as shown in the formula below: .

[0090] Finally, the following constraints are set: In the same newly added same-city order, the pickup and delivery nodes must be accessed by the same delivery vehicle, and the constraint formulas are as follows: The delivery vehicle is only allowed to perform delivery work after the pickup task is completed, and the constraint is as follows: Constraints This represents time-related constraints used to ensure that the actual delivery time does not exceed the service time window limit of the delivery node corresponding to the new same-city order; constraint formula This constraint represents the load capacity of delivery vehicles at pickup nodes, and stipulates that the load capacity of a delivery vehicle leaving any pickup node should not be less than the pickup quantity at that node (the corresponding new pickup and delivery demand), and should not exceed the maximum load capacity of the delivery vehicle; (Constraint formula) This represents the constraint on the load capacity of delivery vehicles at each receiving node, and also constrains that the load capacity of a delivery vehicle leaving any receiving node must not exceed the vehicle's maximum load capacity minus the delivery volume at that receiving node (the corresponding new pickup and delivery demand). Here, N represents the number of orders to be delivered, including e-commerce orders and local orders; V represents the node set, V={0,1,2,…,n,n+1,…,2n}, where 0 represents the target center node; E represents the edge set, E={(i,j)}, i,j∈V; V + V represents the set of pickup nodes. + ={0,1,2,…,n}, V - V represents the set of pickup nodes. - ={ n+1,…,2n};V d V represents the set of delivery nodes. d ={1,2,…,n,n+1,…,2n};K represents the set of delivery vehicles, K={1,2,…,k};c1 represents the fixed usage cost per unit vehicle;c2 represents the unit distance cost of delivery vehicles; Indicate whether to use the k-th delivery vehicle; if yes, ,otherwise, d ij x represents the distance from the i-th delivery node to the j-th delivery node, where i, j ∈ V; ijk Indicates whether the k-th delivery vehicle has traveled from the i-th delivery node to the j-th delivery node. If so, x ijk =1, otherwise, x ijk =0; [A i Bi ] represents the service time window of the i-th delivery node, A i B represents the earliest arrival time of the delivery vehicle agreed upon at the i-th delivery node. i This represents the latest arrival time of the delivery vehicle agreed upon at the i-th delivery node; [C] i D i ] represents the soft service time window of the i-th delivery node, C i B represents the earliest arrival time of a delivery vehicle that the i-th delivery node can tolerate. i This represents the latest arrival time that the i-th delivery node can tolerate for a delivery vehicle; W represents the maximum cargo capacity of the delivery vehicle; M is a set positive number; θ1 indicates that the delivery vehicle arrives earlier than A. i The waiting penalty coefficient corresponding to the time reaching the corresponding delivery node; θ2 represents the delivery vehicle being later than B. i The delay penalty coefficient corresponding to the time when the corresponding delivery node is reached; v represents the speed of the delivery vehicle; Sik represents the service time of the k-th delivery vehicle serving the i-th delivery node, i∈V,t ijk This represents the time it takes for the k-th delivery vehicle to travel from the i-th delivery node to the j-th delivery node; i L represents the quantity of goods at the i-th delivery node. ik T represents the total volume of goods after serving the i-th delivery node. ik Let represent the time when the k-th delivery vehicle starts service at the i-th delivery node, where i∈V and k∈K.

[0091] In this embodiment, the insertion operation will cause a change in cost. The cost change includes not only the increased distance cost brought by the new same-city order, but also the penalty cost caused by violating the time window constraint, as well as the penalty cost change caused by the delivery of subsequent orders. For example, if the new same-city order l (including pickup node l1 and delivery node l2) can be inserted between i1, j1 and i2, j2 respectively, then the resulting cost change can be calculated as follows:

[0092] First, based on the calculation results, the insertion of new same-city orders will increase the vehicle's travel distance, so a path-saving operator is used. The additional transportation costs are calculated by determining the changes in the delivery vehicle's route, which are then used as the formula: .

[0093] Furthermore, when delivery vehicles receive notification of new orders, they will quickly head to the node location corresponding to the new delivery order. However, it cannot be guaranteed that they will arrive on time and begin providing service within the specified time window. Considering the high timeliness requirements for new same-city orders, in this embodiment, new same-city orders that exceed the new service time window are rejected from being inserted into the candidate route segment of the existing first delivery route. Instead, new vehicles are selected for delivery. The corresponding cost calculation formula is as follows: 0.

[0094] Finally, when a new same-city order is inserted into the currently executing delivery route, it will cause a delay in the time it takes for vehicles to arrive at subsequent delivery nodes. This delay may result in some delivery nodes failing to receive delivery within the time window, thus incurring penalty costs. However, the service time window for some cross-border e-commerce express orders may be met due to the delay, which helps reduce the previous waiting penalty costs. Therefore, it is necessary to calculate the changes in the penalty costs of delivery nodes in both of these aspects. To consider the insertion costs of both pickup and delivery nodes simultaneously, analysis shows that, apart from the pickup node needing to be inserted before the delivery node, the two do not affect each other in terms of other costs. Therefore, it is only necessary to consider minimizing the total cost of simultaneously inserting pickup and delivery nodes. The relevant cost calculation formula is as follows:

[0095]

[0096]

[0097]

[0098] in, This indicates the length of time that the delivery time of subsequent delivery nodes is delayed. When the insertion condition cannot be met, i.e. At this point, the cost of punishment is 0.

[0099] Based on the above formula, the formula for calculating the total cost of inserting a new same-city order l (including pickup node l1 and delivery node l2) between i1, j1 and i2, j2 respectively is as follows: By calculating the total cost of the planned path corresponding to each candidate path in the candidate list, the planned path that minimizes the cost of the changes can be determined as the final target path segment.

[0100] Through steps 61 to 63 above, the ALNS-RIM hybrid optimization algorithm is used to perform insertion planning based on the selected candidate path segments, so as to generate the optimal target path segment, provide data for outputting the optimal second delivery information, improve the efficiency of joint cooperation, and reduce delivery costs.

[0101] This embodiment also provides a device for dynamic joint delivery route optimization of e-commerce orders and same-city orders. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0102] Figure 4 This is a structural block diagram of the dynamic joint delivery route optimization device for e-commerce orders and same-city orders according to an embodiment of this application, such as... Figure 4 As shown, the device includes an acquisition module 41, an insertion module 42, and a processing module 43, wherein,

[0103] The acquisition module 41 is used to acquire the first joint delivery information currently being executed when a delivery request for a newly added same-city order is received. The first joint delivery information is generated by using the ALNS-RIM hybrid optimization algorithm, which is composed of real-time insertion method and adaptive large-domain search algorithm, to dynamically optimize the joint delivery path of historical joint delivery information and the previous newly added same-city order.

[0104] The insertion module 41, coupled to the acquisition module 42, is used to acquire all current first delivery routes from the first joint delivery information, and to use the RIM of the ALNS-RIM hybrid optimization algorithm to detect first delivery routes into which new same-city orders can be inserted, thereby obtaining candidate delivery routes, and to detect the candidate route segments corresponding to the insertion of new same-city orders in the candidate delivery routes.

[0105] The processing module 43, coupled to the insertion module 42, is used to perform insertion path planning based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, generate target path segments, and update all first delivery paths according to the target path segments to generate path optimization results including second joint delivery information, wherein the second joint delivery information includes the target delivery path corresponding to the target delivery segment and the unupdated first delivery path.

[0106] In some embodiments, the insertion module 42 further includes:

[0107] The first acquisition unit is used to acquire the order information of each new same-city order from the delivery request. The order information includes the new pickup and delivery node pair corresponding to the new same-city order. The new pickup and delivery node pair is used to represent the pickup node and delivery node corresponding to the new same-city order.

[0108] The first calculation unit, coupled to the first acquisition unit, is used to determine the pickup node and receiving node of the new pickup and delivery node pair corresponding to each new same-city order, and to determine the distance between the pickup node and receiving node corresponding to the same new same-city order and the target center node, thereby obtaining the first path distance and the second path distance.

[0109] The first determining unit, coupled to the first calculating unit, is used to take the longest path distance between the first path distance and the second path distance as the target distance, and after determining the target node corresponding to the target distance, to determine a preset area based on the target node and the target distance, wherein the center of the preset area is located at the target node position.

[0110] The first selection unit, coupled to the first determination unit, is used to select the first delivery path within the preset area from all first delivery paths to obtain candidate delivery paths.

[0111] In some embodiments, the order information also includes the corresponding pickup node and the newly added service time window corresponding to the receiving node, and the insertion module 42 further includes:

[0112] The second acquisition unit is used to obtain the new service time window of the pickup node and the receiving node corresponding to the new same-city order from the order information, and to detect the undelivered first delivery node from all the first delivery nodes corresponding to each candidate delivery path, and to determine the path segment where all the undelivered first delivery nodes are located from the candidate delivery path, so as to obtain the candidate path segment.

[0113] The second selection unit, coupled to the second acquisition unit, is used to select alternative path segments from all candidate path segments after determining the service time window corresponding to each first delivery node on the candidate path segment, based on the time difference between the corresponding newly added service time window and each service time window.

[0114] The second calculation unit, coupled to the second selection unit, is used to determine the available load capacity of the corresponding delivery vehicle based on the total pickup and delivery demand corresponding to the alternative route segment and the current load capacity of the corresponding delivery vehicle. The current load capacity is used to represent the remaining load capacity of the delivery vehicle after loading the goods of the undelivered first delivery node.

[0115] The second determining unit, coupled to the second calculation unit, is used to select, from the alternative path segments, alternative path segments whose carrying capacity is not less than the new pickup and delivery demand obtained from the order information, to obtain candidate path segments. The new pickup and delivery demand is used to characterize the amount of goods collected from the pickup node or delivered to the receiving node.

[0116] In some embodiments, the second selection unit is further configured to determine whether all service time windows corresponding to the candidate path segment exceed the new service time window based on the time difference between each corresponding service time window and the new service time window; and if it is determined that none of the service time windows corresponding to the candidate path segment exceed the new service time window, the candidate path segment is determined as a candidate path segment.

[0117] In some embodiments, before obtaining candidate route segments, the optimization device is further configured to select new delivery vehicles based on the new pickup and delivery demand corresponding to the corresponding new same-city order, if the available capacity is less than the new pickup and delivery demand or if it is determined that at least one service time window exceeds the new service time window; generate a new delivery route corresponding to the corresponding new same-city order using the ALNS of the ALNS-RIM hybrid optimization algorithm, at least with the pickup node and delivery node corresponding to the corresponding new same-city order as the corresponding delivery node; and generate second joint delivery information by combining the new delivery route and all first delivery routes.

[0118] In some embodiments, the processing module 43 further includes:

[0119] The first detection unit is used to detect the first delivery path associated with the target delivery path in all first delivery paths, and obtain the delivery path to be updated.

[0120] The first update unit, coupled to the first detection unit, is used to determine the candidate path segments for each delivery path to be updated, and update the corresponding candidate path segments to the target path segments to obtain the target delivery path.

[0121] The first generation unit, coupled to the first update unit, is used to generate second joint delivery information from the target delivery route and the unupdated first delivery route.

[0122] In some embodiments, the processing module 43 further includes:

[0123] The third determining unit is used to determine all first delivery nodes corresponding to the candidate path segment, and to determine the pickup node and receiving node corresponding to at least one new same-city order associated with each candidate path segment, so as to obtain a delivery node set, wherein the delivery nodes in the delivery node set include all first delivery nodes, all pickup nodes and all receiving nodes.

[0124] The first planning unit, coupled to the third determining unit, is used to process the delivery node set according to a preset path planning method to generate candidate path data. The path planning method includes one of the following: using ALNS of the ALNS-RIM hybrid optimization algorithm to perform neighborhood search iteration on the delivery nodes of the delivery node set; or using RIM of the ALNS-RIM hybrid optimization algorithm to perform node insertion traversal.

[0125] The first processing unit, coupled to the first planning unit, is used to determine the variable cost corresponding to the planned path corresponding to the candidate path data according to a preset objective function, and select the planned path with the minimum variable cost from all planned paths corresponding to the candidate path data to obtain the target path segment. The objective function is constructed based on the increased delivery cost of delivery vehicles and the late delivery penalty cost caused by violating the time window. The delivery distance corresponding to the increased delivery cost of delivery vehicles is determined based on the latitude and longitude of the corresponding delivery node.

[0126] In some embodiments, the first planning unit is further configured to use the RIM of the ALNS-RIM hybrid optimization algorithm to determine two adjacent first delivery nodes located on the candidate path segment, and set the interval between the two adjacent first delivery nodes as the interval to be inserted; randomly insert the pickup node and the receiving node corresponding to the same new same-city order into at least one interval to be inserted, and generate a first initial path, wherein the pickup node is inserted before the receiving node; repeatedly insert at least one of the pickup node and the receiving node in all intervals to be inserted to obtain multiple first initial paths, and use the multiple first initial paths as candidate path data.

[0127] In some embodiments, the first planning unit is further configured to construct paths for all delivery nodes in the delivery node set according to a preset construction method, generating multiple second initial paths, wherein in the second initial paths, the pickup node is located before the receiving node, and the construction method includes one of the following: random insertion method, nearest insertion method, and greedy insertion method; using a roulette wheel algorithm, a target destruction operation and a target repair operation are selected from multiple destruction operations and multiple repair operations corresponding to ALNS in the ALNS-RIM hybrid optimization algorithm, wherein... The destroy operation represents the removal of delivery nodes from the second initial path. The destroy operation includes one of the following: random removal, worst-case removal, and similar removal. The repair operation represents the insertion of the removed delivery nodes into the second initial path after removal. The repair operation includes one of the following: random insertion and greedy insertion. The target repair operation is performed on the reconstructed path obtained by performing the target destroy operation on the second initial path to generate a third path. Based on the objective function, the variable cost corresponding to the third path is determined, and the destroy and repair operations are iterated on the third path according to the variable cost until a fourth path with a variable cost less than a preset cost threshold is generated, thus obtaining candidate path data, where the candidate path data includes all fourth paths.

[0128] This embodiment also provides a service platform, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0129] Optionally, the service platform may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0130] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0131] S1. When receiving a delivery request for a newly added same-city order, obtain the first joint delivery information currently being executed. The first joint delivery information is generated by using the ALNS-RIM hybrid optimization algorithm, which is composed of real-time insertion method and adaptive large-domain search algorithm, to dynamically optimize the joint delivery path of historical joint delivery information and the previous newly added same-city order.

[0132] S2. Obtain all current first delivery routes from the first joint delivery information, and use the RIM of the ALNS-RIM hybrid optimization algorithm to detect first delivery routes into which new same-city orders can be inserted, obtain candidate delivery routes, and detect the candidate route segments corresponding to the insertion of new same-city orders in the candidate delivery routes.

[0133] S3. Based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, insert path planning is performed to generate target path segments. Based on the target path segments, all first delivery paths are updated to generate path optimization results including second joint delivery information. The second joint delivery information includes the target delivery path corresponding to the target delivery segment and the unupdated first delivery path.

[0134] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0135] Furthermore, in conjunction with the dynamic joint delivery route optimization method for e-commerce orders and same-city orders in the above embodiments, this application embodiment can provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the dynamic joint delivery route optimization methods for e-commerce orders and same-city orders in the above embodiments.

[0136] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0137] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for optimizing the dynamic joint delivery route of e-commerce orders and same-city orders, characterized in that, include: When a delivery request for a newly added same-city order is received, the first joint delivery information currently being executed is obtained. The first joint delivery information is generated by using the ALNS-RIM hybrid optimization algorithm, which is composed of real-time insertion method and adaptive large-domain search algorithm, to dynamically optimize the joint delivery path of historical joint delivery information and the previous newly added same-city order. From the first joint delivery information, obtain all current first delivery routes, and using the RIM of the ALNS-RIM hybrid optimization algorithm, detect the first delivery routes into which the new same-city order can be inserted to obtain candidate delivery routes, and detect the candidate path segments corresponding to the insertion of the new same-city order from the candidate delivery routes; obtain the order information of each new same-city order from the delivery request, wherein the order information includes the new pickup and delivery node pair corresponding to the new same-city order, and the new pickup and delivery node pair is used to represent the pickup node and delivery node corresponding to the new same-city order; determine the corresponding route segment for each new same-city order. The process involves adding a pickup node and a receiving node to a new pickup and delivery node pair, and determining the distances between the pickup node and the receiving node corresponding to the same new same-city order and the delivery center, thus obtaining a first path distance and a second path distance. The longest path distance between the first path distance and the second path distance is taken as the target distance. After determining the target node corresponding to the target distance, a preset area is determined based on the target node and the target distance, wherein the center of the preset area is located at the target node. From all the first delivery paths, the first delivery paths located within the preset area are selected to obtain the candidate delivery path segments. Based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, insertion path planning is performed to generate target path segments. Based on the target path segments, all first delivery paths are updated to generate path optimization results including second joint delivery information. The second joint delivery information includes the target delivery path corresponding to the target delivery segment and the unupdated first delivery path.

2. The optimization method according to claim 1, characterized in that, The order information also includes the corresponding new service time windows for the pickup node and the receiving node. In the candidate delivery routes, the candidate route segment corresponding to the newly added same-city order is detected and inserted, including: From the order information, obtain the new service time window of the pickup node and the receiving node corresponding to the new same-city order, and detect the undelivered first delivery node from all the first delivery nodes corresponding to each candidate delivery path, and determine the path segment where all the undelivered first delivery nodes are located from the candidate delivery path to obtain the candidate path segment; After determining the service time window corresponding to each of the first delivery nodes on the candidate route segment, alternative route segments are selected from all the candidate route segments based on the time difference between the corresponding new service time window and each service time window. Based on the total pickup and delivery demand corresponding to the alternative route segment and the current cargo capacity of the corresponding delivery vehicle, the available cargo capacity of the corresponding delivery vehicle is determined, wherein the current cargo capacity is used to represent the remaining cargo capacity of the delivery vehicle after loading the goods of the first delivery node that have not been delivered. Among the alternative route segments, the alternative route segments whose carrying capacity is not less than the new pickup and delivery demand obtained from the order information are selected to obtain the candidate route segments, wherein the new pickup and delivery demand is used to characterize the amount of goods collected from the pickup node or the amount of goods delivered to the receiving node.

3. The optimization method according to claim 2, characterized in that, Based on the newly added service time window and the time difference of each service time window, candidate path segments are selected from all the candidate path segments, including: Based on the time difference between each of the corresponding service time windows and the newly added service time window, determine whether all the service time windows corresponding to the candidate path segment exceed the newly added service time window; If it is determined that none of the service time windows corresponding to the candidate path segment have exceeded the newly added service time window, the candidate path segment is determined as the alternative path segment.

4. The optimization method according to claim 3, characterized in that, Before obtaining the candidate path segment, the method further includes: If the available carrying capacity is less than the new pickup and delivery demand or if at least one of the service time windows is determined to exceed the new service time window, a new delivery vehicle is selected based on the new pickup and delivery demand corresponding to the new same-city order. At least the pickup node and the receiving node corresponding to the newly added same-city order are used as the corresponding delivery nodes. The ALNS of the ALNS-RIM hybrid optimization algorithm is used to generate a new delivery route corresponding to the newly added same-city order. The new delivery route and all of the first delivery routes are used to generate the second joint delivery information.

5. The optimization method according to claim 2, characterized in that, Based on the target path segment, all the first delivery paths are updated, including: Among all the first delivery routes, the first delivery route associated with the target delivery route is detected to obtain the delivery route to be updated; Determine the candidate path segments for each of the delivery routes to be updated, and update the corresponding candidate path segments to the target path segments to obtain the target delivery route; The target delivery route and the unupdated first delivery route are used to generate the second joint delivery information.

6. The optimization method according to claim 2, characterized in that, Based on the ALNS-RIM hybrid optimization algorithm and the corresponding candidate path segments, insertion path planning is performed to generate target path segments, including: All first delivery nodes corresponding to the candidate path segment are determined, and the pickup node and the receiving node corresponding to at least one new same-city order associated with each candidate path segment are determined to obtain a delivery node set, wherein the delivery nodes of the delivery node set include all first delivery nodes, all pickup nodes and all receiving nodes; The delivery node set is processed according to a preset path planning method to generate candidate path data. The path planning method includes one of the following: using the ALNS of the ALNS-RIM hybrid optimization algorithm to perform neighborhood search iteration on the delivery nodes in the delivery node set; or using the RIM of the ALNS-RIM hybrid optimization algorithm to perform node insertion traversal. According to a preset objective function, the variable cost corresponding to the planned path corresponding to the candidate path data is determined, and the planned path with the smallest variable cost is selected from all the planned paths corresponding to the candidate path data to obtain the target path segment. The objective function is constructed based on the increased delivery cost of delivery vehicles and the late delivery penalty cost caused by violating the time window. The delivery distance corresponding to the increased delivery cost of delivery vehicles is determined based on the latitude and longitude of the corresponding delivery node.

7. The optimization method according to claim 6, characterized in that, The delivery node set is processed according to a preset route planning method to generate candidate route data, including: Using the RIM of the ALNS-RIM hybrid optimization algorithm, two adjacent first delivery nodes located on the candidate path segment are determined, and the interval between the two adjacent first delivery nodes is set as the interval to be inserted. The pickup node and the receiving node corresponding to the same new same-city order are randomly inserted into at least one of the insertion intervals, and a first initial path is generated, wherein the pickup node is inserted before the receiving node; Repeatedly insert at least one of the pickup node and the receiving node in all the intervals to be inserted to obtain multiple first initial paths, and use the multiple first initial paths as the candidate path data.

8. The optimization method according to claim 6, characterized in that, The delivery node set is processed according to a preset route planning method to generate candidate route data, including: According to a preset construction method, paths are constructed for all delivery nodes in the delivery node set to generate multiple second initial paths, wherein the pickup node is located before the receiving node in the second initial path, and the construction method includes one of the following: random insertion method, nearest insertion method, and greedy insertion method; Using the roulette wheel algorithm, a target destruction operation and a target repair operation are selected from a variety of destruction operations and a variety of repair operations corresponding to ALNS in the ALNS-RIM hybrid optimization algorithm. The destruction operation represents the removal of the delivery node in the second initial path, and the destruction operation includes one of the following: random removal, worst-case removal, and similar removal. The repair operation represents the insertion of the removed delivery node into the second initial path after the removal is completed, and the repair operation includes one of the following: random insertion and greedy insertion. For the reconstructed path obtained by performing the target destroy operation on the second initial path, perform the target repair operation to generate the third path; Based on the objective function, the variable cost corresponding to the third path is determined, and the third path is iterated through destroy and repair operations according to the variable cost until a fourth path with a variable cost less than a preset cost threshold is generated, thereby obtaining the candidate path data, wherein the candidate path data includes all the fourth paths.

9. A service platform, comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the steps of the method for dynamic joint delivery route optimization of e-commerce orders and same-city orders as described in any one of claims 1 to 8.

Images