Order delivery path planning method, system, server and storage medium

Abstract

The application provides an order delivery path planning method and system, a server and a storage medium, relates to the technical field of logistics, solves the problems that the traditional decision scheme cannot balance the working time of delivery personnel and cannot avoid the detour of delivery personnel, and comprises the following steps: according to the data of a collection and distribution station, time distance matrix data and delivery personnel data, solving a first path planning strategy of a planable order based on a preset first path planning target, determining a target delivery personnel and the to-be-delivered orders of the target delivery personnel; according to real-time time distance matrix data and real-time position data, solving a second path planning strategy of the to-be-delivered orders based on a preset second path planning target, and determining real-time delivery path planning information of the target delivery personnel. The application realizes intelligent matching of delivery personnel and planable orders and intelligent decision of delivery personnel and delivery paths, and improves the delivery efficiency of delivery personnel.

Description

Technical Field

[0001] This application relates to the field of logistics technology, specifically to an order delivery route planning method, system, server, and storage medium. Background Technology

[0002] The Vehicle Routing Problem (VRP) is a classic problem in the logistics field. Common project solutions include assigning staff based on experience, developing scheduling models or heuristic algorithms based on the scenario, and academically developing precise algorithms such as the branch-and-price algorithm and the branch-and-cut-and-price algorithm.

[0003] Currently, the typical logistics and delivery method involves sending an order to a nearby courier point before the customer receives it, at which point the courier initiates the delivery. To improve the customer shopping experience, some companies select multiple distribution centers within a city, each covering different urban areas. Each distribution center is staffed with multiple couriers as needed. Before the customer receives the order, it no longer passes through courier points but is instead centralized at its respective distribution center, where couriers deliver it uniformly.

[0004] For any distribution center, there is a problem that the distribution of delivery orders and couriers cannot be evenly distributed. When couriers are waiting to deliver orders, they cannot take into account the real-time road conditions, which leads to detours and long delivery times. Summary of the Invention

[0005] This application provides an order delivery route planning method, system, server, and storage medium that solves the problems of existing route planning methods, such as the inability to decide the number of delivery personnel, the inability to balance the working hours of delivery personnel, and the inability to prevent delivery personnel from taking detours. It can optimize the matching of delivery orders and delivery personnel and better determine the delivery routes of delivery personnel.

[0006] On the one hand, this application provides an order delivery route planning method, including:

[0007] Obtain the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of delivery personnel to be planned;

[0008] Based on the data from the distribution station, the time-distance matrix data, and the delivery personnel data, and based on a preset first path planning target, the first path planning strategy for the plannable orders is solved to determine at least one of the delivery personnel to be planned as the target delivery personnel and at least one of the plannable orders as the delivery orders of the target delivery personnel.

[0009] Obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered;

[0010] Based on the real-time time distance matrix data and the real-time location data, and based on the preset second path planning target, the second path planning strategy for the order to be delivered is solved to determine the real-time delivery path planning information of the target delivery person.

[0011] In one possible implementation of this application, obtaining the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of the delivery personnel to be planned includes:

[0012] Obtain multiple initial orders that contain at least one of the planarable orders;

[0013] The current distribution station address in the distribution station data and the initial order addresses in the multiple initial orders are analyzed to obtain the latitude and longitude information of the current distribution station and the latitude and longitude information of the multiple initial order addresses;

[0014] The Cartesian product is performed on the union of the latitude and longitude information of the current distribution station and the latitude and longitude information of multiple initial order addresses to obtain multiple first flow direction data between the current distribution station address and multiple initial order addresses, as well as between different initial order addresses;

[0015] Based on multiple first flow direction data, flow direction verification is performed on the multiple initial orders to determine at least one of the plannable orders among the multiple initial orders and the time distance matrix data between the at least one of the plannable orders.

[0016] In one possible implementation of this application, the step of solving the first path planning strategy for the plannable orders based on a preset first path planning target, and determining at least one of the planned delivery personnel as target delivery personnel and determining at least one of the plannable orders as orders to be delivered to the target delivery personnel, includes:

[0017] Based on the objectives of minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time, the first route planning strategy for the plannable order is solved to obtain the first delivery route planning decision result for the plannable order;

[0018] Based on the first delivery route planning decision result, at least one of the delivery personnel to be planned is identified as the target delivery personnel, and at least one of the plannable orders is identified as the delivery orders to be delivered by the target delivery personnel.

[0019] In one possible implementation of this application, solving the first route planning strategy for the plannable order to obtain the first delivery route planning decision result for the plannable order includes:

[0020] With minimizing the number of delivery personnel as the initial objective, an initial solution for path planning that satisfies multiple preset path planning constraints is generated.

[0021] Set the initial solution of the path planning as the current solution and the optimal solution of the path planning;

[0022] Based on the current solution of the path planning, a neighborhood search is iterated, and the current solution of the path planning and the optimal solution of the path planning are updated until the neighborhood search iteration stops;

[0023] The optimal solution of the route planning obtained after the update is used as the first delivery route planning decision result of the plannable order.

[0024] In one possible implementation of this application, the step of performing neighborhood search iterations based on the initial solution of the path planning, and updating the current solution of the path planning and the optimal solution of the path planning until the neighborhood search iterations stop, includes:

[0025] Based on the current solution of the path planning, a neighborhood search is performed to obtain a neighborhood solution of the path planning that satisfies multiple path planning constraints.

[0026] The neighborhood solution of the path planning is scored according to multiple preset path planning scoring functions to obtain the neighborhood solution scoring result;

[0027] Based on the neighborhood solution scoring results, the current solution and the optimal solution of the path planning are updated;

[0028] Repeat the above steps until the neighborhood search iteration stops.

[0029] In one possible implementation of this application, obtaining the real-time location data of the target delivery person and the real-time time-distance matrix data of the order to be delivered includes:

[0030] Obtain real-time delivery route planning trigger commands;

[0031] Based on the real-time delivery route planning trigger instruction, the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the order to be delivered are obtained.

[0032] In one possible implementation of this application, after obtaining the real-time delivery route planning trigger instruction, the method further includes:

[0033] Based on the real-time delivery route planning trigger instruction, determine whether all the orders to be delivered corresponding to the target delivery person have been delivered.

[0034] If not all of the orders to be delivered are completed, the orders to be delivered for the target delivery personnel shall be updated.

[0035] When all the orders to be delivered are completed, the order delivery route planning for the target delivery personnel ends.

[0036] In one possible implementation of this application, obtaining the real-time location data of the target delivery person and the real-time time-distance matrix data of the order to be delivered includes:

[0037] The real-time location address in the real-time location data and the address of the order to be delivered in the order to be delivered are analyzed to obtain the real-time location latitude and longitude information and at least one address latitude and longitude information of the order to be delivered;

[0038] The union of the real-time location latitude and longitude information and the latitude and longitude information of at least one of the orders to be delivered is processed by Cartesian product to obtain multiple second flow direction data between the real-time location address and at least one of the orders to be delivered, as well as between different orders to be delivered;

[0039] Based on multiple second flow direction data, the flow direction of the orders to be delivered is verified to determine the real-time time distance matrix data between the orders to be delivered.

[0040] In one possible implementation of this application, the step of solving the second path planning strategy for the order to be delivered based on a preset second path planning target to determine the real-time delivery path planning information of the target delivery person includes:

[0041] Based on the objective of minimizing the number of order deliveries, an initial solution for real-time path planning that satisfies multiple preset real-time path planning constraints is generated.

[0042] Set the initial solution of the real-time path planning as the current solution and the optimal solution of the real-time path planning;

[0043] The neighborhood search is iterated based on the current solution of the real-time path planning, and the current solution and the optimal solution of the real-time path planning are updated until the neighborhood search iteration stops.

[0044] The optimal solution for real-time route planning obtained after the update is completed will be used as the real-time delivery route planning information for the target delivery personnel.

[0045] In one possible implementation of this application, the step of performing neighborhood search iterations based on the current solution of real-time path planning, and updating the current solution and the optimal solution of real-time path planning until the neighborhood search iterations stop, includes:

[0046] Based on the current solution of the real-time path planning, a neighborhood search is performed to obtain a real-time path planning neighborhood solution that satisfies multiple real-time path planning constraints.

[0047] The real-time path planning neighborhood solution is scored according to multiple preset real-time path planning scoring functions to obtain the real-time neighborhood solution scoring result.

[0048] Based on the real-time neighborhood solution scoring results, the current solution and the optimal solution of the real-time path planning are updated.

[0049] Repeat the above steps until the neighborhood search iteration stops.

[0050] On the other hand, this application provides an order delivery route planning system, the system comprising:

[0051] The first acquisition module is used to acquire the current distribution station data, the time-distance matrix data of the plannable orders, and the delivery personnel data of the delivery personnel to be planned.

[0052] The first route planning module is used to solve the first route planning strategy for the plannable orders based on the distribution station data, the time distance matrix data and the delivery personnel data, and a preset first route planning target, to determine at least one of the delivery personnel to be planned as the target delivery personnel and to determine at least one of the plannable orders as the delivery orders of the target delivery personnel.

[0053] The second acquisition module is used to acquire the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the order to be delivered;

[0054] The second path planning module is used to solve the second path planning strategy for the order to be delivered based on the real-time time distance matrix data and the real-time location data, and on a preset second path planning target, to determine the real-time delivery path planning information of the target delivery personnel.

[0055] On the other hand, this application also provides a server, the server comprising:

[0056] One or more processors;

[0057] Memory; and

[0058] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the order delivery route planning method.

[0059] On the other hand, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program being loaded by a processor to perform the steps in the order delivery route planning method.

[0060] This application, based on current distribution station data, time-distance matrix data of plannable orders, and delivery personnel data of planned delivery personnel, and based on a preset first path planning objective, solves the first path planning strategy for plannable orders, identifying at least one planned delivery personnel as target delivery personnel and at least one plannable order as a delivery order for a target delivery personnel, thereby optimizing the number of target delivery personnel deployed and balancing the number of delivery orders delivered by each target delivery personnel, thus balancing the working hours of all deployed delivery personnel; during the real-time delivery process of target delivery personnel, based on the real-time time-distance matrix data and real-time location data of target delivery personnel, and based on a preset second path planning objective, the second path for the delivery orders is determined. The path planning strategy is used to solve the problem and determine the real-time delivery route planning information of the target delivery personnel. That is, based on the orders to be delivered and the real-time location of the target delivery personnel, the delivery route of the target delivery personnel is planned in real time to reduce the possibility of the target delivery personnel taking detours. Therefore, the delivery route planning decision result obtained by the above scheme can meet the two indicator requirements of reducing the detours of delivery personnel and balancing the working time of all delivery personnel. It solves the problems of traditional decision-making schemes, such as the inability to decide the number of delivery personnel, the inability to balance the working time of delivery personnel, and the inability to avoid detours of delivery personnel. This application realizes the intelligent matching of delivery personnel and plannable orders, and realizes the intelligent decision-making of delivery personnel and delivery routes for plannable orders, thereby improving the delivery efficiency of delivery personnel. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a schematic diagram of the action node scenario of the delivery personnel provided in the embodiments of this application;

[0063] Figure 2 This is a schematic diagram of a delivery order scenario provided in the embodiments of this application;

[0064] Figure 3 This is a schematic flowchart of an embodiment of the order delivery route planning method provided in this application.

[0065] Figure 4 This is a schematic diagram of an embodiment of the order pool model provided in this application.

[0066] Figure 5 This is a schematic flowchart of an embodiment of performing a neighborhood search on the current solution provided in this application.

[0067] Figure 6 This is a schematic diagram of a scenario where a neighborhood search is performed on the current solution, as provided in an embodiment of this application.

[0068] Figure 7 This is a schematic diagram of an embodiment of the order delivery route planning system provided in this application.

[0069] Figure 8 This is a schematic diagram of an embodiment of the server provided in this application. Detailed Implementation

[0070] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by creative effort are within the scope of protection of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0072] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0073] Currently, the typical logistics and delivery method involves sending an order to a nearby courier point before the customer receives it, at which point a delivery person initiates the delivery. To improve the customer shopping experience, some companies select multiple distribution centers within a city, each covering a different area. Each distribution center is staffed with multiple delivery personnel as needed. Before the customer receives the order, it no longer passes through courier points but is instead centralized at its designated distribution center, where delivery personnel handle the entire process. For example... Figure 1 As shown, for any distribution station, the action nodes of the delivery personnel at the distribution station can be the following steps: After starting work, the delivery personnel first sort out the orders they are responsible for delivering, and then start delivery. The delivery personnel arrive at the order address corresponding to each order one by one, deliver the order to the customer, and go off duty on the spot after all orders are delivered.

[0074] Delivery personnel's working hours are usually related to the number of orders to be delivered and the order addresses that day. Since the number of orders and the order addresses change every day, it is difficult to guarantee the average working hours of each delivery person if a fixed delivery area is assigned to them.

[0075] The current business model presents several scenarios: The number of delivery personnel on duty at the distribution center each day is limited. When the number of orders is small, all delivery personnel on duty that day are assigned delivery tasks. When the number of orders is large, the number of delivery personnel on duty that day cannot guarantee the timely completion of all orders, so other delivery personnel are temporarily called in to complete the order tasks. After receiving all the orders to be delivered, the delivery personnel manually decide the order delivery order, which is difficult to consider in terms of real-time road conditions and can easily lead to detours. There are scenarios where delivery personnel transfer orders among themselves after receiving the orders to be delivered. There are also scenarios where orders cannot be received by the customer on the first delivery of the day due to customer reasons, requiring a second delivery on the same day or delivery the next day.

[0076] Currently, existing decision-making solutions aim to average the workload of each delivery person at the distribution center by optimizing metrics such as the number of delivery orders and working hours. However, these solutions may encounter unreasonable scenarios when optimizing these metrics. For example, ... Figure 2 As shown, delivery person A has only one long-distance order to deliver, while delivery person B has multiple short-distance orders concentrated in one area. In this scenario, delivery person A and delivery person B may have the same working hours. If the goal is to average the number of orders delivered by each delivery person, delivery person A would need to return to the area with concentrated short-distance orders after completing the long-distance order to deliver another 1-2 orders, while delivery person B would remain in the area with concentrated short-distance orders but could only deliver 2-3 orders. Obviously, the latter does not conform to the actual business operation scenario.

[0077] Delivery personnel typically need to minimize detours when delivering orders. Avoiding detours means minimizing the total working time for all delivery personnel. However, if the model only focuses on averaging the working time of delivery personnel, it will cause the model to allocate orders that involve detours to achieve an average, which in turn will increase the total working time of delivery personnel.

[0078] Furthermore, in the distribution station business scenario involved in this application, delivery personnel can be divided into two categories: fixed-duty delivery personnel and temporarily deployed delivery personnel. According to business requirements, if fixed-duty delivery personnel can complete all orders for the day before the stipulated time, then there is no need to deploy temporary delivery personnel, and the workload of fixed-duty delivery personnel is averaged. If fixed-duty delivery personnel cannot complete all orders for the day before the stipulated time, then temporary delivery personnel need to be deployed appropriately, and the workload of all deployed delivery personnel is averaged. Clearly, the decision on whether to deploy temporary delivery personnel and how many to deploy also depends on the daily order volume of the distribution station, and must be made before the "goods sorting" time node of the distribution station.

[0079] Since delivery personnel's income is linked to their workload, and considering the average workload of delivery personnel, it is necessary to ensure that the working hours of regular delivery personnel are relatively balanced when no temporary delivery personnel are called in, and to ensure that the working hours of both regular and temporary delivery personnel are relatively balanced when temporary delivery personnel are called in.

[0080] Therefore, determining the number of delivery personnel based on business requirements, while minimizing detours, balancing the working hours of all personnel, and ultimately assigning each delivery person their assigned orders to achieve matching between personnel and orders, enabling delivery personnel to sort orders after starting work, and addressing issues such as unavoidable road congestion during real-time order delivery after sorting, cannot yield high-quality decisions by relying on traditional methods based on human experience.

[0081] To address the aforementioned issues, this application provides an order delivery route planning method, system, server, and storage medium, which will be described in detail below.

[0082] like Figure 3 and Figure 4 The diagram shown is a flowchart of an embodiment of the order delivery route planning method in this application. The order delivery route planning method includes steps 101 to 104:

[0083] 101. Obtain the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of delivery personnel to be planned.

[0084] In practical applications, logistics and distribution companies typically set up multiple distribution stations in a city based on business requirements. Each distribution station covers different areas of that city. Before a customer receives an order, the order is centralized at the distribution station in the area to which the order belongs, and then delivered uniformly by the delivery personnel at that distribution station. This application describes any one of the multiple distribution stations; in this embodiment, that distribution station is defined as the current distribution station.

[0085] The current distribution station data may include the current distribution station name, current distribution station address, and current distribution station business hours, and may also include other data, which is not specifically limited here; multiple initial order data are related data of orders belonging to the current distribution station, which is not specifically limited here.

[0086] The delivery personnel to be planned include both fixed and temporary delivery personnel belonging to the current distribution station. Delivery personnel data may include the names, numbers, and main delivery areas of fixed and temporary delivery personnel, as well as other data, which are not specifically limited here.

[0087] In this embodiment, the orders received by the current distribution station each day are defined as initial orders. Since the initial orders received by the current distribution station each day are not fixed, some of these initial orders can be planned, while others are unplanned and require dedicated personnel for delivery. Therefore, before making delivery route planning decisions each day, it is necessary to obtain the time distance information between the current distribution station and the order addresses in the initial orders, filter out the plannable orders from multiple initial orders, and parse the flow direction data between the plannable orders. This flow direction data between the plannable orders is then used as the time distance matrix data for the plannable orders.

[0088] The current distribution station's data and the delivery personnel data for planned deliveries are both stored on the current distribution station's server. When a new order is assigned to the current distribution station, the server can record multiple initial order data related to that order. During application, this data can be obtained by directly calling it from the server or by manual input. This embodiment does not specifically limit the method of obtaining the distribution station data, the time-distance matrix data of planned orders, and the delivery personnel data.

[0089] 102. Based on the data from the distribution station, the time-distance matrix data, and the delivery personnel data, and based on the preset first path planning target, solve the first path planning strategy for the plannable orders, determine at least one delivery personnel to be planned as the target delivery personnel, and determine at least one plannable order as the target delivery personnel for the order to be delivered.

[0090] In this embodiment, the delivery route planning decision can be determined through a model. Specifically, the model determines at least one planned delivery person as the target delivery person and at least one plannable order as the delivery order to be delivered to the target delivery person. Therefore, this application defines the multiple delivery orders allocated to the target delivery person as the delivery order pool for the target delivery person, and the model used to determine the deployment of target delivery persons and the allocation of delivery orders to them as the order pool model. In practice, the order pool model also determines the delivery route for each delivery person. During application, a neighborhood search algorithm is used to solve the order pool model to determine the delivery route planning decision.

[0091] Specifically, in this embodiment, such as Figure 4 As shown, the data of the distribution station, the data of the planarable orders, the time distance matrix data of the planarable orders, and the data of the delivery personnel are used as the input data of the order pool model. Based on the neighborhood search algorithm, the first path planning strategy of the planarable orders is solved to obtain at least one delivery personnel to be planned as the target delivery personnel and at least one planarable order as the target delivery personnel, that is, the order pool of the target delivery personnel is obtained.

[0092] 103. Obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered.

[0093] After the server of the current distribution station determines the target delivery personnel and the orders to be delivered by the target delivery personnel, it allocates multiple orders to be delivered to the target delivery personnel, forming a corresponding order pool on the target delivery personnel's end. In the "sorting goods" stage, the target delivery personnel of the current distribution station sort out the corresponding orders to be delivered according to the order pool and deliver the orders.

[0094] The real-time location data of the target delivery personnel refers to their current location during the delivery process. This data can be represented using geographic text information, coordinate data, or other forms of data. It can be uploaded by the target delivery personnel themselves or obtained remotely from a server. This embodiment does not impose specific limitations on this.

[0095] By analyzing the real-time location data of the target delivery personnel and the address data of the orders to be delivered, the real-time time-distance matrix data of the orders to be delivered is obtained, which is the flow direction data between the orders to be delivered.

[0096] Because the real-time location data of the target delivery personnel and the real-time order data in the order pool are constantly changing—for example, the transfer of orders between target delivery personnel and the removal of orders from the order pool after delivery—the order pool is subject to real-time changes. Therefore, during the delivery process, it is necessary to obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of different orders to be delivered, and then perform route planning based on these data.

[0097] The real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered are both stored in the server of the current distribution station. When the target delivery personnel are delivering orders to be delivered, the server can correspondingly record the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered. In the application process, the above data can be obtained by directly calling the above data from the server, or it can be obtained by manually inputting the above data. This embodiment does not specifically limit the method of obtaining the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered.

[0098] 104. Based on the real-time time distance matrix data and real-time location data, and based on the preset second path planning target, solve the second path planning strategy for the order to be delivered, and determine the real-time delivery path planning information of the target delivery personnel.

[0099] In this embodiment, the real-time delivery route planning information of the target delivery personnel can be determined using a model. This application defines the model used to determine the real-time delivery route planning information of the target delivery personnel as a real-time route planning model, and uses a neighborhood search algorithm to solve the real-time route planning model to determine the real-time delivery route planning information of the target delivery personnel.

[0100] Specifically, in this embodiment, real-time time distance matrix data and real-time location data are used as input data for the real-time path planning model. Based on the neighborhood search algorithm, the second path planning strategy for the order to be delivered is solved to obtain the real-time delivery path planning information of the target delivery personnel. The real-time delivery path planning information of the target delivery personnel is then sent to the target delivery personnel's terminal in real time to guide the target delivery personnel to deliver orders according to the delivery path planning information, thereby reducing the situation of delivery personnel taking detours and reducing the working time of delivery personnel.

[0101] Therefore, this application can optimize the number of target delivery personnel and balance the number of orders to be delivered by each target delivery personnel through the above-mentioned scheme, thereby balancing the working time of all deployed delivery personnel; and based on the orders to be delivered and the real-time location of the target delivery personnel, the delivery route of the target delivery personnel is planned in real time, reducing the possibility of the target delivery personnel taking detours; therefore, the delivery route planning decision result obtained by adopting the above scheme can meet the two indicator requirements of reducing the detours of delivery personnel and balancing the working time of all deployed delivery personnel, solving the problems of traditional decision-making schemes that cannot determine the number of delivery personnel to be deployed, cannot balance the working time of delivery personnel, and cannot avoid the detours of delivery personnel. This application realizes intelligent matching of delivery personnel and plannable orders, and realizes intelligent decision-making of delivery personnel and delivery routes for plannable orders, thereby improving the delivery efficiency of delivery personnel.

[0102] In this application, since an order pool model is used to determine the target delivery personnel and assign orders to them, and a real-time route planning model is used to determine the real-time delivery route planning information for the target delivery personnel, multiple different data variables input into the order pool model and the real-time route planning model need to be defined to facilitate decision-making. For example, the following definitions apply to these multiple different data variables in this application:

[0103] Current distribution station: C;

[0104] Current distribution station address: add c ;

[0105] The current number of delivery personnel to be planned at the distribution center is: cqty;

[0106] Initial order set: O, where o∈O, o represents an initial order in the initial order set O;

[0107] Initial order address: add o ;

[0108] The current initial order volume for the distribution center today: oqty;

[0109] Delivery time for orders pending delivery: t o ;

[0110] Latest delivery time for orders awaiting dispatch: duet = latest delivery time (set according to actual needs);

[0111] Unplannable order set: NPSO;

[0112] Plannable Order Set: PSO;

[0113] Set of delivery personnel to be planned: I, where i∈I, i represents a delivery person in the set of delivery personnel to be planned. Since the delivery personnel to be planned include fixed-duty delivery personnel and temporary delivery personnel, and there is no limit to the number of temporary delivery personnel, the target number of delivery personnel deployed by the current collection and distribution station on the current day is initialized to iqty=oqty, that is, it is assumed that the target number of delivery personnel deployed is equal to the number of initial orders, that is, one delivery person is arranged to deliver each initial order;

[0114] Target delivery personnel working hours: t i ;

[0115] Maximum working hours for target delivery personnel: max t = maximum working hours (set according to actual needs), target delivery personnel working time = delivery time for pending orders + customer service time. The working time of target delivery personnel can also include lunch break, evening break, waiting time for customer signature, etc., which can be set according to actual needs.

[0116] Target delivery personnel's mileage: d i ;

[0117] Maximum mileage for target delivery personnel: max d = maximum mileage (set according to actual needs);

[0118] Whether the target delivery personnel are involved in the decision variable: x i x i∈{0,1}. In this example, the multiple different data variables in this application include, but are not limited to, the above-mentioned methods. In addition, other definition methods can be used to define the multiple different data variables in this application as needed. This embodiment does not specifically limit this.

[0119] In another embodiment of this application, obtaining the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of the delivery personnel to be planned specifically includes the following steps 201-204:

[0120] 201. Obtain multiple initial orders that contain at least one plannable order;

[0121] In this embodiment, obtaining multiple initial orders is equivalent to obtaining multiple initial order data corresponding to multiple initial orders. After the current distribution station receives multiple initial orders within its delivery range each day, the initial order data of the multiple initial orders is saved to the server of the current distribution station. The multiple initial order data may include information about the distribution station, order number, order delivery address, order recipient, and order recipient contact information, and may also include other data. This embodiment does not specifically limit this.

[0122] 202. Parse the current distribution station address and the initial order addresses in the multiple initial orders from the distribution station data to obtain the latitude and longitude information of the current distribution station and the multiple initial order addresses. In this embodiment, a latitude and longitude query function is used to parse the latitude and longitude coordinates of the current distribution station address and the multiple initial order addresses to obtain the latitude and longitude information of the current distribution station and the multiple initial order addresses. The latitude and longitude query function can be a GeoHash index function, or other functions or algorithms, and this application does not specifically limit it.

[0123] During the parsing process, since the latitude and longitude information of the current distribution station is fixed, parsing will not fail. However, multiple initial orders may fail to be parsed due to insufficient detail in the initial order address. Therefore, in this embodiment, initial orders with obtained initial order address latitude and longitude information are considered plannable orders, while initial orders without obtained initial order address latitude and longitude information are considered unplannable orders.

[0124] In this embodiment, for ease of understanding, the set of planned orders stored in the server is defined as the planned order set PSO, and the set of unplanned orders stored in the server is defined as the unplanned order set NPSO.

[0125] 203. Perform Cartesian product processing on the union of the current distribution station's latitude and longitude information and the latitude and longitude information of multiple initial order addresses to obtain multiple first flow direction data between the current distribution station address and multiple initial order addresses, as well as between different initial order addresses.

[0126] For example, performing a Cartesian product on the union of n initial order address latitude and longitude information and 1 current distribution station latitude and longitude information will yield (n+1)*(n+1) first flow direction data.

[0127] 204. Based on multiple first flow direction data, perform flow direction verification on multiple initial orders to determine at least one plannable order among the multiple initial orders and the time distance matrix data between at least one plannable order;

[0128] Specifically, the navigation software is called to analyze each first flow data obtained in step 203 to obtain navigation distance information and navigation time information from the flow start address to the flow destination address;

[0129] In this embodiment, since some of the first flow data obtained in step 203 may fail to be acquired, it is necessary to perform flow verification on multiple initial order data to further determine the plannable and unplannable orders of multiple initial order data.

[0130] Specifically, the process takes the initial order address to be verified as the starting address and determines whether at least one flow from the initial order address to be verified to another initial order address has been successfully obtained. The process also takes the initial order address to be verified as the destination address and determines whether at least one flow from another initial order address to the initial order address to be verified has been successfully obtained.

[0131] Based on step 202, that is, after the initial orders with obtained initial order address latitude and longitude information are placed into the plannable order set PSO, the initial orders in the plannable order set PSO that satisfy the requirement of successfully obtaining both bidirectional flow are retained in the plannable order set PSO, and the initial orders in the plannable order set PSO that do not satisfy the requirement of successfully obtaining both bidirectional flow are transferred to the unplannable order set NPSO.

[0132] In this embodiment, all initial orders in the final planarable order set PSO are considered planarable orders, and all initial orders in the final unplanarable order set NPSO are considered unplanarable orders.

[0133] In this application, the planable orders in the planable order set PSO are delivered using the current distribution station's business model, and the unplanable orders in the unplanable order set NPSO are delivered using a dedicated delivery method. This embodiment does not specifically limit the delivery method for unplanable orders.

[0134] By going through steps 201 to 204 above, the plannable orders and the time distance matrix data corresponding to the plannable orders can be obtained. In this embodiment, the time distance matrix data corresponding to the plannable orders is the first flow direction data between multiple order addresses in the plannable orders.

[0135] In another embodiment of this application, such as Figure 5 As shown, the first path planning objective is set as minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time. Therefore, based on the preset first path planning objective, the first path planning strategy for plannable orders is solved to determine at least one delivery person to be planned as the target delivery person and at least one order to be planned as the target delivery person, including:

[0136] Based on the objectives of minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time, the first route planning strategy for plannable orders is solved to obtain the first delivery route planning decision results for plannable orders;

[0137] Based on the results of the first delivery route planning decision, at least one delivery person to be planned is identified as the target delivery person, and at least one order to be planned is identified as the target delivery person's order to be delivered.

[0138] Specifically, the first route planning strategy for plannable orders is solved to obtain the first delivery route planning decision results for plannable orders, including:

[0139] With minimizing the number of delivery personnel as the initial objective, an initial solution for path planning that satisfies multiple preset path planning constraints is generated. The initial solution is then set as the current solution and the optimal solution for path planning. Neighborhood search iteration is performed based on the current solution, and the current solution and the optimal solution are updated until the neighborhood search iteration stops. The optimal solution obtained after the update is completed is used as the first delivery path planning decision result for the plannable order data.

[0140] The initial solution is the first solution set when performing a neighborhood search using the neighborhood search algorithm. In the neighborhood search initialization of this embodiment, the goal is to minimize the number of delivery personnel involved, and the solution that satisfies multiple path planning constraints is the first solution, i.e., the initial solution of path planning.

[0141] Depend on Figure 5 As can be seen, when performing the neighborhood search algorithm, a neighborhood search is performed on a current solution to obtain multiple new neighborhood solutions, and then a better solution is selected from the multiple new neighborhood solutions to update the current solution; in the neighborhood search initialization of this embodiment, the initial solution of path planning is used as the current solution of path planning;

[0142] The optimal solution is the best solution obtained during the neighborhood search iteration. The optimal solution is the solution that best satisfies the objectives of minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time among all solutions obtained from the order pool model. In the neighborhood search initialization of this embodiment, the initial solution of the path planning is taken as the optimal solution of the path planning. After the neighborhood search iteration stops, the updated optimal solution of the path planning is taken as the first delivery path planning decision result of the plannable orders.

[0143] In this embodiment, a neighborhood search is iterated based on the initial solution of the path planning, and the current solution and the optimal solution of the path planning are updated until the neighborhood search iteration stops, which is equivalent to executing the process in multiple loops. Figure 5 The flowchart may specifically include the following steps 301-304:

[0144] 301. Perform a neighborhood search based on the current solution of the path planning to obtain a neighborhood solution of the path planning that satisfies multiple path planning constraints.

[0145] Specifically, a neighborhood search algorithm is used to perform a neighborhood search on the current solution of the path planning to obtain a neighborhood solution of the path planning that satisfies multiple path planning constraints.

[0146] The steps of this embodiment are described in detail below using a set of examples.

[0147] Specifically, such as Figure 6 As shown, assume that the current distribution station has a planned number of orders of k+m.

[0148] First, when the neighborhood search algorithm solves the order pool model, the initial objective is to minimize the number of delivery personnel. This yields an initial solution for path planning that satisfies multiple preset path planning constraints. In this embodiment, the delivery personnel mentioned below are the delivery personnel to be planned, and the orders mentioned below are the plannable orders. Therefore, the obtained initial solution for path planning can be as follows: Figure 6 As shown in 6-1, specifically:

[0149] Currently, the distribution station has two delivery personnel.

[0150] Delivery person A's route planning strategy is to deliver order 1, order 2, ..., order k in sequence;

[0151] Delivery person B's route planning strategy is to deliver orders k+1, k+2, ..., k+m in sequence.

[0152] In the initial stage of neighborhood search, the above initial solution for path planning is taken as the current solution and the optimal solution for path planning.

[0153] The neighborhood search algorithm iterates through the current solution of the path planning above to obtain a neighborhood solution that satisfies multiple path planning constraints. The obtained neighborhood solution can be as follows: Figure 6 The path planning neighborhood solution shown in Figure 6-2 is 1 and so on. Figure 6 The path planning neighborhood solution 2 is shown in 6-3.

[0154] Specifically, such as Figure 6 As shown in Figure 6-2, the path planning neighborhood solution 1 is based on the current path planning solution. It moves order 1 from delivery person A to delivery person B and sets order 1 for delivery before order k+1. The specific path planning neighborhood solution 1 obtained is as follows:

[0155] Currently, the distribution station has two delivery personnel.

[0156] Delivery person A's route planning strategy is to deliver orders 2, ..., k in sequence;

[0157] Delivery person B's route planning strategy is to deliver order 1, order k+1, order k+2, ..., order k+m in sequence.

[0158] like Figure 6 As shown in 6-3, the path planning neighborhood solution 2 is based on the current path planning solution. By swapping delivery person A's order 1 with delivery person B's order k+1, the specific path planning neighborhood solution 2 is as follows:

[0159] Currently, the distribution station has two delivery personnel.

[0160] Delivery person A's route planning strategy is to deliver order k+1, order 2, ..., order k in sequence;

[0161] Delivery person B's route planning strategy is to deliver order 1, order k+2, ..., order k+m in sequence.

[0162] The above-described neighborhood search iteration method is only an example. In actual applications, the neighborhood search can be performed between any orders in the current solution of the path planning, or between any delivery personnel in the current solution of the path planning. Furthermore, any strategy can be used for neighborhood search, as long as the obtained path planning neighborhood solution satisfies multiple path planning constraints.

[0163] Therefore, this embodiment uses a neighborhood search algorithm to perform neighborhood search iteration based on the current solution of the path planning. The resulting neighborhood solutions of the path planning are diverse. This application does not specifically limit the neighborhood search method for iterating the current solution of the path planning, and it can be set arbitrarily as needed.

[0164] In this embodiment, before performing a neighborhood search based on the current solution of the path planning, the order delivery path planning method further includes: configuring multiple path planning constraints;

[0165] Before using the neighborhood search algorithm to solve the order pool model, multiple path planning constraints in the order pool model are pre-configured. These multiple path planning constraints include, but are not limited to, at least one of the following constraints:

[0166] Constraint 1: ∑ i∈I x i ≥cqty means the total number of target delivery personnel deployed ∑ i∈I x i The number of planned delivery personnel currently available at the distribution station is greater than or equal to cqty.

[0167] Constraint 2: d i ≤maxd, This represents the mileage d traveled by any delivery person. i The mileage d not exceeding the maximum driving distance maxd, and without the deployment of delivery personnel. i =0;

[0168] Constraint 3: t i ≤maxt means that the working time t of any delivery person is t. i The working time shall not exceed the maximum working time maxt, and the working time of personnel not involved in delivery shall be 0.

[0169] Constraint 4: t o ≤duet, This represents the delivery time t of any order. o The delivery time shall not exceed the latest duet time for this order.

[0170] The following examples illustrate the application scenarios of the constraints in this embodiment.

[0171] In this embodiment, for example, the constraints are set as follows: the current distribution station has 1 or 2 fixed on-duty delivery personnel, the maximum driving distance maxd can be set to maxd = 250 kilometers, the maximum working time maxt can be set to maxt = 690 minutes, and the latest delivery time of the order duet can be set to duet = 20:00 in the evening.

[0172] If set Figure 6 The solution shown in 6-2 in [reference] is a neighborhood solution for path planning that satisfies the above-mentioned multiple path planning constraint conditions. This path planning neighborhood solution indicates that:

[0173] The number of delivery personnel invested in the current distribution station is 2, which satisfies the first constraint condition mentioned above;

[0174] The path planning strategy of delivery personnel A is to deliver orders k+1, order 2, …, order k in sequence. And the driving mileage of delivery personnel A after serving order k is less than or equal to 250 kilometers, the working time does not exceed 690 minutes, and the service end time does not exceed 20:00. That is, the path planning strategy of this delivery personnel A satisfies the second, third, and fourth constraint conditions mentioned above;

[0175] The path planning strategy of delivery personnel B is to deliver orders 1, order k+2, …, order k+m in sequence. Similarly, the driving mileage of delivery personnel B after serving order k is also less than or equal to 250 kilometers, the working time does not exceed 690 minutes, and the service end time does not exceed 20:00. That is, the path planning strategy of this delivery personnel B also satisfies the second, third, and fourth constraint conditions mentioned above.

[0176] 302. According to a preset multiple path planning scoring function, score the path planning neighborhood solution to obtain a neighborhood solution scoring result.

[0177] During the neighborhood search iteration process using the neighborhood search algorithm, it is necessary to evaluate the quality of the multiple path planning neighborhood solutions obtained by neighborhood search through multiple path planning scoring functions to obtain the corresponding neighborhood solution scoring results. In this embodiment, it can be set that the smaller the score in the neighborhood solution scoring result, the better the path planning neighborhood solution.

[0178] Exemplarily, if the multiple path planning neighborhood solutions obtained by neighborhood search include path planning neighborhood solution 1 and path planning neighborhood solution 2, use multiple path planning scoring functions to score path planning neighborhood solution 1 and path planning neighborhood solution 1. If the score of path planning neighborhood solution 1 is less than the score of path planning neighborhood solution 2, score(1) < score(2), then the neighborhood solution scoring result is that path planning neighborhood solution 1 is better than path planning neighborhood solution 2.

[0179] The main purpose of this application is to obtain an optimal path planning solution that simultaneously satisfies the requirements of "determining the number of delivery personnel, reducing detours by delivery personnel, and balancing the working time of all delivery personnel." However, a single scoring function cannot simultaneously quantify the three indicators: the number of delivery personnel, detours, and the average working time of delivery personnel. If these three indicators are weighted and standardized, the values ​​of the weighting parameters are difficult to estimate, and these values ​​directly affect the evaluation of the neighborhood solution by the neighborhood search algorithm, thus influencing the selection of the optimal path planning solution.

[0180] Therefore, in order to avoid parameter weighting and tuning based on the above three indicators, a multi-level scoring function is used as the path planning scoring function in this embodiment.

[0181] In this embodiment, the multi-level path planning scoring function specifically includes, but is not limited to, the following path planning scoring functions:

[0182] The first-level path planning scoring function is specifically: ∑ i∈I x i ;

[0183] Where i represents the delivery person's tag, and x i Let I be the i-th delivery person, and I be the total number of delivery people at the current distribution station. i∈I x i The number of delivery personnel deployed in the neighborhood solution of the path planning;

[0184] The second-level path planning scoring function is: max(τ, max error(A));

[0185] maxerror(A)=max(dbs(x1-avg(A)),abs(x2-avg(A))...,abs(x m -avg(A))),

[0186] avg(A)=∑ i∈I x i *t i / ∑ i∈I x i ;

[0187] Among them, t i Let ∑ be the working hours of the i-th delivery person. i∈I x i *t i Let A be the total working time of all delivery personnel involved in the path planning neighborhood solution, avg(A) be the average working time of all delivery personnel in the path planning neighborhood solution, and abs(x) be the average working time of all delivery personnel in the path planning neighborhood solution.m -avg(A) represents the working hours t of the m-th delivery person. m The deviation time from the average working time avg(A) is the deviation time, maxerror(A) is the maximum deviation time among m deviation times, τ is the threshold for the maximum deviation value of working time, and max(τ,max error(A)) is the maximum value after comparing the maximum deviation time with the threshold for the maximum deviation value of working time;

[0188] The third-level path planning scoring function is specifically: ∑ i∈I x i *t i ;

[0189] Where, ∑ i∈I x i *t i This represents the total working time of all delivery personnel involved in the path planning neighborhood solution.

[0190] Correspondingly, the neighborhood solution scoring results include first-level neighborhood solution scoring results, second-level neighborhood solution scoring results, and third-level neighborhood solution scoring results.

[0191] Therefore, in this embodiment, the neighborhood solutions of the path planning are scored according to multiple preset path planning scoring functions to obtain neighborhood solution scoring results, specifically including:

[0192] Based on the first-level, second-level, and third-level path planning scoring functions described above, the neighborhood solutions for path planning are scored to obtain the corresponding neighborhood solution scoring results. Specifically, this includes the following steps 401–404:

[0193] 401. Based on the first-level path planning scoring function, perform a first-level scoring on the path planning neighborhood solution to obtain the first-level neighborhood solution scoring result.

[0194] Since the initial solution for the first route planning strategy of the planable orders was generated in the initial stage with the goal of minimizing the number of delivery personnel, the initial solution for route planning was generated. That is, the goal of minimizing the number of delivery personnel and multiple route planning constraints were determined in the initial stage.

[0195] Therefore, a first-level path planning scoring function is used to score the path planning neighborhood solution, that is, to evaluate the number of delivery personnel deployed in the path planning neighborhood solution. This allows the neighborhood search algorithm to judge the quality of the path planning neighborhood solution based on the already determined number of delivery personnel deployed and the number of delivery personnel deployed in the first-level neighborhood solution scoring results. This avoids the neighborhood search algorithm accepting the path planning neighborhood solution with an increased number of delivery personnel as the optimal solution, and facilitates further balancing of the working time of each delivery person and reducing the detours of delivery personnel.

[0196] 402. After obtaining the first-level neighborhood solution score, the second-level path planning neighborhood solution is scored according to the second-level path planning score function to obtain the second-level neighborhood solution score.

[0197] Based on the first-level neighborhood solution scoring results, a second-level path planning scoring function is used to perform a second-level scoring on the path planning neighborhood solutions. This involves evaluating the maximum deviation of delivery personnel's working time from the average working time in the path planning neighborhood solutions. The second-level path planning scoring function sets a threshold τ for the maximum deviation of working time. This allows the neighborhood search algorithm to judge the quality of the path planning neighborhood solutions based on the deviation of each delivery personnel's working time in the second-level neighborhood solution scoring results during the iteration process. This facilitates the neighborhood search algorithm in selecting the path planning neighborhood solution that balances the working time of each delivery personnel as the optimal solution.

[0198] For example, the path planning neighborhood solutions are set to include path planning neighborhood solution A and path planning neighborhood solution B. Since the second-level path planning scoring function sets a maximum deviation threshold τ for working time, when comparing the merits of solutions during the iteration of the neighborhood search algorithm, as long as the second-level neighborhood solution scoring results of solutions A and B are both less than the maximum deviation threshold τ for working time, it is determined that path planning neighborhood solutions A and B have the same merits in the second-level path planning scoring function stage.

[0199] In this embodiment, the second-level path planning scoring function sets a maximum deviation threshold τ for working time, so that the neighborhood search algorithm optimizes the average working time of delivery personnel to meet business needs. The maximum deviation threshold τ for working time can be set as needed according to different business scenarios to achieve the average working time of delivery personnel, while also leaving room for optimization of the total working time of delivery personnel (without detours).

[0200] 403. After obtaining the second-level neighborhood solution score, the path planning neighborhood solution is scored in three levels according to the third-level path planning score function to obtain the third-level neighborhood solution score.

[0201] Based on the second-level neighborhood solution scoring results, a third-level path planning scoring function is used to perform a three-level scoring on the path planning neighborhood solution. This involves evaluating the total working time of all delivery personnel in the path planning neighborhood solution. During the iteration process of the neighborhood search algorithm, the merits of the path planning neighborhood solution are judged based on the total working time of all delivery personnel in the three-level neighborhood solution scoring results. This facilitates the neighborhood search algorithm in selecting the path planning neighborhood solution with the shortest total working time of delivery personnel as the optimal solution, thereby preserving the optimal solution where delivery personnel do not take detours.

[0202] 404. The third-level neighborhood solution score is taken as the final neighborhood solution score.

[0203] Since the three-level neighborhood solution scoring result is the score obtained after the path planning neighborhood solution has been evaluated by the first-level path planning scoring function, the second-level path planning scoring function, and the third-level path planning scoring function, the three-level neighborhood solution scoring result is used as the final neighborhood solution scoring result. This facilitates the neighborhood search algorithm to iteratively select the path planning neighborhood solution that best satisfies the requirements of "deciding on the number of delivery personnel, reducing detours for delivery personnel, and balancing the working time of all delivery personnel" as the optimal solution.

[0204] For example, when the path planning neighborhood solution includes two solutions, such as path planning neighborhood solution A and path planning neighborhood solution B, the first-level path planning scoring function is used to determine the merits of path planning neighborhood solution A and path planning neighborhood solution B. If they cannot be distinguished, the second-level path planning scoring function is used to determine the merits of path planning neighborhood solution A and path planning neighborhood solution B. If they still cannot be distinguished, the third-level path planning scoring function is used to determine the merits of path planning neighborhood solution A and path planning neighborhood solution B, in order to select the path planning neighborhood solution that can simultaneously achieve the above objectives.

[0205] 303. Update the current solution and the optimal solution of the path planning based on the neighborhood solution score.

[0206] Based on the neighborhood solution scoring results, the current solution and the optimal solution of the path planning are updated, specifically including:

[0207] Obtain the receiving rules, which include the first receiving rule and the second receiving rule;

[0208] Based on the first or second receiving rule, the current solution and the optimal solution of the path planning are updated according to the neighborhood solution scoring results.

[0209] In another embodiment of this application, the current solution and the optimal solution of the path planning are updated based on the first receiving rule and the neighborhood solution scoring results.

[0210] In this embodiment, the first receiving rule is as follows: in each loop iteration, the optimal path planning neighborhood solution is selected to update the current path planning solution and the optimal path planning solution. The optimal path planning solution is always updated with the optimal path planning neighborhood solution in multiple loop iterations (the optimal path planning solution records the best solution searched in history).

[0211] In each iteration, the optimal solution of the path planning is evaluated by multiple path planning scoring functions. After obtaining the neighborhood solution scoring results, the optimal neighborhood solution of the path planning is selected from the neighborhood solution scoring results according to the preset first acceptance rule to update the current solution of the path planning. That is, the optimal neighborhood solution of the path planning obtained in the current iteration is used as the current solution and the optimal solution of the path planning in the next iteration. The optimal solution of the path planning only records the best solution found in the current iteration and in the past multiple iterations.

[0212] In another embodiment of this application, the current solution and the optimal solution of the path planning are updated based on the second receiving rule and the neighborhood solution scoring results.

[0213] In this embodiment, the second receiving rule is as follows: during a certain loop iteration, a worse or better path planning neighborhood solution is randomly selected to update the current path planning solution and the optimal path planning solution. The optimal path planning solution is still the best solution found in the current iteration and in previous loop iterations.

[0214] In a random iteration, the optimal solution of the path planning is evaluated by multiple path planning scoring functions. After obtaining the neighborhood solution scoring results, the current solution of the path planning is updated by randomly selecting the inferior or superior path planning neighborhood solution from the neighborhood solution scoring results according to the preset second receiving rule and using the simulated annealing algorithm. That is, the inferior or superior path planning neighborhood solution in the current iteration is randomly selected as the current solution and the optimal solution of the path planning in the next iteration. The optimal solution of the path planning is still the best solution found in the current iteration and in the past multiple iterations.

[0215] 304. Repeat steps 301 to 303 above until the neighborhood search iteration stops.

[0216] In this embodiment, a neighborhood search iteration stop operation is performed according to a preset termination condition.

[0217] The termination condition can be specifically that the execution time of the neighborhood search algorithm exceeds a set time, or that the optimal solution of the path planning is not updated within a set time. After the neighborhood search iteration stops, the optimal solution of the path planning is output as the first delivery path planning decision result.

[0218] After obtaining the first delivery route planning decision result, the server of the current distribution station determines the target delivery personnel and the orders to be delivered by the target delivery personnel based on the first delivery route planning decision result, and assigns the corresponding orders to the multiple target delivery personnel, forming a corresponding order pool at the target delivery personnel's end. In the "sorting goods" stage, the target delivery personnel of the current distribution station sort out the corresponding orders to be delivered according to the order pool and deliver the orders.

[0219] During real-time delivery, the pool of orders to be delivered changes according to the delivery progress, and the real-time road conditions also change in real time for any given delivery person. For example, the changes in the pool of orders to be delivered for a target delivery person may include, but are not limited to, the following:

[0220] (1) Once a customer signs for a pending delivery order, the target delivery personnel will remove the signed pending delivery order from the pending delivery order pool.

[0221] (2) If a customer refuses to accept a delivery order, the target delivery personnel will remove the rejected delivery order from the delivery order pool.

[0222] (3) If the first delivery of an order to be delivered fails on the same day, the customer requests to change the delivery address. The target delivery person modifies the information of the order to be delivered and keeps it in the order pool.

[0223] (4) If the first delivery of an order to be delivered fails on the same day and the customer requests to change the delivery time, if the customer requests delivery at a certain time on the same day, the order to be delivered will still be kept in the order pool of the current target delivery person and the delivery time requirement will be recorded; if the customer requests delivery on a different day, the order to be delivered will be removed from the order pool and the order to be delivered will be subsequently delivered through other channels.

[0224] (5) If the target delivery person A is unable to continue delivering the orders in the order pool due to an emergency, and needs to transfer some or all of the orders to the target delivery person B, the order pool of the target delivery person B will be increased by some or all of the orders of the target delivery person A. At the same time, the target delivery person A will remove the corresponding orders or clear the order pool.

[0225] At the same time, the real-time location of the target delivery personnel and the real-time traffic conditions in their area will also change in real time. Based on these changes, the optimal delivery route of the target delivery personnel based on their real-time location needs to be replanned in real time. Therefore, it is necessary to determine the real-time delivery route planning information of the target delivery personnel based on the real-time location data of the target delivery personnel and the real-time time distance matrix data of multiple orders to be delivered, so as to further plan the optimal delivery route of the target delivery personnel in real time.

[0226] In another embodiment of this application, obtaining real-time location data of the target delivery person and real-time time-distance matrix data of the orders to be delivered includes:

[0227] Obtain the real-time delivery route planning trigger command, and based on the real-time delivery route planning trigger command, obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered.

[0228] In this embodiment, the target delivery person may issue a real-time delivery route planning trigger instruction (hereinafter referred to as instruction) on the target delivery person's terminal (hereinafter referred to as the terminal in this embodiment), or the server of the current collection and distribution station may automatically generate the instruction under certain conditions. After receiving the instruction, the server obtains the real-time location data of the target delivery person and the real-time time distance matrix data of the order to be delivered according to the instruction.

[0229] In this embodiment, the target delivery person can issue instructions from the terminal in any scenario. For example, instructions can be issued in the following scenarios, but are not limited to:

[0230] (1) After the target delivery personnel have completed the sorting of the orders to be delivered;

[0231] (2) After the target delivery person arrives at the location of one of the pending orders and completes the pending order, the pending order is removed from the pending order pool;

[0232] (3) When target delivery personnel A and target delivery personnel B engage in order transfer, target delivery personnel A and target delivery personnel B shall issue instructions on their respective terminals.

[0233] (4) When a customer of one of the pending orders requests a change to the delivery address and delivery time;

[0234] (5) When the target delivery personnel obtain real-time traffic conditions to the location of the order to be delivered and there is congestion, and the delivery route needs to be replanned;

[0235] (6) When the starting point and ending point of the currently planned order delivery route obtained by the target delivery personnel are different from the starting point and ending point of the actual delivery route;

[0236] (7) When the real-time location of the target delivery personnel changes due to their own reasons.

[0237] In this embodiment, the server can also be set to generate an instruction once at a set interval. This embodiment does not specifically limit the acquisition scenario and acquisition method of the real-time delivery route planning trigger instruction.

[0238] In this embodiment, by obtaining instructions, the real-time location data of the target delivery person and the real-time time-distance matrix data of the orders to be delivered are obtained. This allows the target delivery person to trigger the real-time route planning model when needed, making the real-time delivery route planning more flexible. It also allows the target delivery person to obtain the next destination anytime and anywhere based on their real-time location, the real-time order pool, and real-time traffic conditions. This ensures that the next destination the target delivery person travels to is the best choice at the moment, which also makes the delivery route taken by the target delivery person when delivering all the orders to be delivered reasonable.

[0239] In another embodiment of this application, to avoid unnecessary cyclical acquisition of the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered after the target delivery personnel have completed the delivery tasks of all orders to be delivered, in this embodiment, after obtaining the real-time delivery route planning trigger instruction, the order delivery route planning method further includes:

[0240] Based on real-time delivery route planning trigger instructions, determine whether all pending orders corresponding to the target delivery personnel have been delivered, and

[0241] When not all orders to be delivered have been completed, update the orders to be delivered for the target delivery personnel.

[0242] Once all pending orders have been delivered, the order delivery route planning for the target delivery personnel will end.

[0243] That is, when the target delivery person issues a real-time delivery route planning trigger command or the server generates a real-time delivery route planning trigger command, at this time, it is determined whether all the orders to be delivered by the target delivery person have been delivered. If so, the delivery route planning of the target delivery person's orders is terminated, thereby avoiding the server from obtaining the real-time location data of the target delivery person and the real-time time distance matrix data of the orders to be delivered again, so as to avoid resource waste.

[0244] In this embodiment, the methods for determining whether all orders to be delivered by the target delivery person for the day have been completed may include, but are not limited to, the following:

[0245] (1) When a target delivery person finishes delivering an order, the order is removed from the target delivery person's order pool. When all orders in the target delivery person's order pool are deleted, it means that all orders for that target delivery person have been delivered that day.

[0246] (2) When the target delivery person finishes delivering an order, the target delivery person marks the order as delivered. When all the orders in the target delivery person's order pool are marked as delivered, it means that all the orders of the target delivery person for that day have been delivered.

[0247] This embodiment does not specify the method for determining whether all pending orders for the target delivery person have been delivered on a given day.

[0248] In another embodiment of this application, obtaining the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered specifically includes the following steps 501-503:

[0249] 501. Parse the real-time location address in the real-time location data and the address of the order to be delivered in the order to be delivered to obtain the latitude and longitude information of the real-time location and at least one latitude and longitude information of the address of the order to be delivered;

[0250] In this embodiment, a latitude and longitude query function is used to parse the latitude and longitude coordinates of the real-time location address and the latitude and longitude coordinates of multiple order addresses to be delivered, so as to obtain the latitude and longitude information of the real-time location and the latitude and longitude information of at least one order address to be delivered; the latitude and longitude query function can be a GeoHash index function, or other functions or algorithms, and this application does not specifically limit it.

[0251] In this embodiment, since the orders to be delivered have already been successfully parsed using latitude and longitude information, all currently parsed orders to be delivered are parsable. If the real-time location address of the target delivery person fails to be parsed, the target delivery person can manually adjust the real-time location address on their terminal to achieve successful parsing. Therefore, both the real-time location latitude and longitude information and the latitude and longitude information of at least one order to be delivered address can be successfully parsed.

[0252] 502. Perform Cartesian product processing on the union of real-time location latitude and longitude information and at least one order address latitude and longitude information to obtain multiple second flow direction data between real-time location address and at least one order address to be delivered, as well as between different order addresses to be delivered.

[0253] For example, performing a Cartesian product on the union of the latitude and longitude information of n orders to be delivered and the latitude and longitude information of one real-time location will yield (n+1)*(n+1) second flow direction data.

[0254] 503. Based on multiple secondary flow data, perform flow direction verification on the orders to be delivered to determine the real-time time distance matrix data between the orders to be delivered.

[0255] The navigation software is invoked to analyze each second flow direction data obtained in step 502, and to obtain the navigation distance information and navigation time information from the flow direction start address to the flow direction destination address;

[0256] By following the steps above, you can obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered.

[0257] In another embodiment of this application, minimizing the number of order deliveries is set as the second route planning objective. Therefore, based on the preset second route planning objective, the second route planning strategy for the orders to be delivered is solved to determine the real-time delivery route planning information of the target delivery personnel, including:

[0258] Based on the objective of minimizing the number of order deliveries, an initial solution for real-time path planning that satisfies multiple preset real-time path planning constraints is generated; the initial solution for real-time path planning is then set as the current solution and the optimal solution for real-time path planning.

[0259] The neighborhood search is iterated based on the current solution of the real-time path planning, and the current solution and the optimal solution of the real-time path planning are updated until the neighborhood search iteration stops.

[0260] The optimal solution for real-time route planning obtained after the update is completed will be used as the real-time delivery route planning information for the target delivery personnel.

[0261] In this embodiment, the preset real-time route planning constraint includes that each order to be delivered is delivered only once. Therefore, minimizing the number of order deliveries is taken as the initial objective, and the solution that satisfies the route planning constraint that each order to be delivered is delivered only once is taken as the first solution, i.e., the initial solution of real-time route planning.

[0262] In the neighborhood search initialization of this embodiment, the initial solution of path planning is used as the current solution of path planning, and the initial solution of path planning is also used as the optimal solution of path planning. After the neighborhood search iteration stops, the updated optimal solution of path planning is used as the first delivery path planning decision result for the plannable orders. The optimal solution of real-time path planning is the solution that best satisfies the objective of minimizing the number of order deliveries and the path planning constraint that each order to be delivered is delivered only once, among all solutions obtained from the real-time path planning model. In another embodiment of this application, neighborhood search iteration is performed based on the current solution of real-time path planning, and the current solution of real-time path planning and the optimal solution of real-time path planning are updated until the neighborhood search iteration stops. This can specifically include the following steps 601-604:

[0263] 601. Based on the current solution of real-time path planning, perform a neighborhood search to obtain a real-time path planning neighborhood solution that satisfies multiple real-time path planning constraints.

[0264] Specifically, a neighborhood search algorithm is used to perform a neighborhood search on the current solution of the real-time path planning to obtain a neighborhood solution of the real-time path planning that satisfies multiple real-time path planning constraints.

[0265] In practical applications, the neighborhood search can be performed between any orders to be delivered in the current solution of the real-time path planning, and any strategy can be used for the neighborhood search, as long as the obtained real-time path planning neighborhood solution satisfies the real-time path planning constraints.

[0266] Therefore, this embodiment uses a neighborhood search algorithm to perform neighborhood search iteration based on the current solution of real-time path planning. The resulting neighborhood solutions for real-time path planning are diverse. This application does not specifically limit the neighborhood search method for iterating the current solution of real-time path planning, and it can be set arbitrarily as needed.

[0267] In this embodiment, before performing a neighborhood search based on the current solution of the real-time path planning, the order delivery path planning method further includes: configuring multiple real-time path planning constraints;

[0268] Before using the neighborhood search algorithm to solve the order pool model, multiple real-time route planning constraints in the real-time route planning model are pre-configured. In this embodiment, the real-time route planning constraints include that each order to be delivered is delivered only once. Alternatively, according to different real-time delivery route planning requirements, the driving mileage constraints and working hours constraints of the target delivery personnel can be set. This embodiment does not make specific limitations on this.

[0269] 602. Based on multiple preset real-time path planning scoring functions, score the real-time path planning neighborhood solutions to obtain the real-time neighborhood solution scoring results.

[0270] During the iteration of neighborhood search using the neighborhood search algorithm, it is necessary to evaluate the merits of multiple real-time path planning neighborhood solutions obtained by the neighborhood search through multiple real-time path planning scoring functions to obtain the corresponding real-time neighborhood solution scoring results. In this embodiment, based on the real-time neighborhood solution scoring results, the optimal solution among all real-time path planning neighborhood solutions is selected.

[0271] In practical applications, customers can modify the delivery time after the first delivery failure of a real-time order. If the time is still the same day, the real-time route planning model will add or subtract a certain duration from that delivery time to configure it as the delivery time window for the current real-time order. If the current real-time order is not delivered before the delivery time plus the specified duration, the real-time route planning model considers the current real-time order to be late.

[0272] To prevent delivery personnel from being late during the delivery process, the optimal solution for real-time route planning can be obtained by setting a real-time route planning scoring function for the real-time route planning model.

[0273] Therefore, in this embodiment, in order to meet the above-mentioned requirement of avoiding lateness, a multi-level real-time path planning scoring function is adopted. Specifically, the real-time path planning model presets multiple real-time path planning scoring functions, including but not limited to the following:

[0274] Level 1 Real-Time Route Planning Scoring Function: Minimizes the number of late orders awaiting delivery;

[0275] Second-level real-time route planning scoring function: Minimize the total delay time of late orders;

[0276] The third-level real-time route planning scoring function aims to minimize the total delivery time for the target delivery personnel.

[0277] Correspondingly, the real-time neighborhood solution scoring results include first-level real-time neighborhood solution scoring results, second-level real-time neighborhood solution scoring results, and third-level real-time neighborhood solution scoring results.

[0278] Therefore, in this embodiment, the real-time path planning neighborhood solution is scored according to multiple preset real-time path planning scoring functions to obtain the real-time neighborhood solution scoring result, specifically including:

[0279] Based on the first-level, second-level, and third-level real-time path planning scoring functions mentioned above, the real-time path planning neighborhood solutions are scored to obtain the corresponding real-time neighborhood solution scoring results.

[0280] Specifically, the steps 701-704 are included:

[0281] 701. Based on the first-level real-time path planning scoring function, perform a first-level scoring on the real-time path planning neighborhood solution to obtain the first-level real-time neighborhood solution scoring result.

[0282] To prioritize customer satisfaction with order delivery timeliness, the first-level real-time route planning scoring function of the real-time route planning model allows for late orders to be delivered, with the minimum number of late orders. This is because when the time windows of real-time orders to be delivered in the order pool are concentrated in the same time period, a single delivery person cannot guarantee on-time delivery of all these orders at the same time. In this case, delivery is carried out according to the principle of minimizing the number of late orders to ensure delivery satisfaction.

[0283] 702. After obtaining the first-level real-time neighborhood solution score, the real-time path planning neighborhood solution is scored in a second-level manner according to the second-level real-time path planning score function to obtain the second-level real-time neighborhood solution score.

[0284] Based on the first-level real-time neighborhood solution scoring results, a second-level real-time path planning scoring function is used to perform a second-level scoring on the path planning neighborhood solution. The second-level real-time path planning scoring function minimizes the total time of all late-delivery orders for the purpose of avoiding excessive loss of order timeliness in special scenarios.

[0285] 703. After obtaining the second-level real-time neighborhood solution score, the real-time path planning neighborhood solution is scored at the third level according to the third-level real-time path planning score function to obtain the third-level real-time neighborhood solution score.

[0286] The third-level real-time route planning scoring function aims to reduce the working time of delivery personnel, i.e., to reduce detours, while avoiding excessive losses in order delivery time. The neighborhood search algorithm uses these three real-time route planning scoring functions to select the optimal real-time route planning neighborhood solution from multiple solutions, and then designates this optimal solution as the best real-time route planning solution.

[0287] For example, when the real-time path planning neighborhood solution includes two solutions, such as real-time path planning neighborhood solution A and real-time path planning neighborhood solution B, the first-level real-time path planning scoring function is used to determine the merits of real-time path planning neighborhood solution A and real-time path planning neighborhood solution B. If they cannot be distinguished, the second-level real-time path planning scoring function is used to determine the merits of real-time path planning neighborhood solution A and real-time path planning neighborhood solution B. If they still cannot be distinguished, the third-level real-time path planning scoring function is used to determine the merits of real-time path planning neighborhood solution A and real-time path planning neighborhood solution B, so as to select the optimal solution among real-time path planning neighborhood solution A and real-time path planning neighborhood solution B.

[0288] 603. Update the current solution and the optimal solution of real-time path planning based on the real-time neighborhood solution score.

[0289] Based on the real-time neighborhood solution scoring results, the optimal real-time path planning solution is recorded from the real-time path planning neighborhood solutions of multiple iterations. The current real-time path planning solution and the optimal real-time path planning solution are updated. Based on the optimal real-time path planning solution, the real-time delivery path planning information of the target delivery personnel is determined, thereby guiding the next vehicle route of the target delivery personnel and realizing the real-time planning of the target delivery personnel's delivery path.

[0290] 604. Repeat steps 601 to 603 above until the neighborhood search iteration stops.

[0291] To better implement the order delivery route planning method in the embodiments of this application, an order delivery route planning system 800 is also provided in the embodiments of this application, such as... Figure 7 As shown, the order delivery route planning system 800 includes:

[0292] The first acquisition module 801 is used to acquire the current distribution station data, the time-distance matrix data of the plannable orders, and the delivery personnel data of the delivery personnel to be planned.

[0293] The first route planning module 802 is used to solve the first route planning strategy for plannable orders based on the data of the distribution station, the time distance matrix data and the delivery personnel data, and the preset first route planning target, to determine at least one delivery personnel to be planned as the target delivery personnel and at least one plannable order as the target delivery personnel.

[0294] The second acquisition module 803 is used to acquire the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered;

[0295] The second path planning module 804 is used to solve the second path planning strategy for the order to be delivered based on the real-time time distance matrix data and real-time location data and the preset second path planning target, and to determine the real-time delivery path planning information of the target delivery personnel.

[0296] The first acquisition module 801 is further specifically as follows:

[0297] Used to obtain multiple initial orders containing at least one plannable order;

[0298] This is used to parse the current distribution station address and the initial order addresses from multiple initial orders in the distribution station data, and obtain the latitude and longitude information of the current distribution station and the latitude and longitude information of multiple initial order addresses;

[0299] This is used to perform Cartesian product processing on the union of the current distribution station's latitude and longitude information and the latitude and longitude information of multiple initial order addresses, to obtain multiple first flow direction data between the current distribution station address and multiple initial order addresses, as well as between different initial order addresses;

[0300] This is used to perform flow direction verification on multiple initial orders based on multiple first flow direction data, and to determine at least one plannable order among the multiple initial orders and the time distance matrix data between at least one plannable order.

[0301] The first path planning module 802 is further specifically as follows:

[0302] This method is used to solve the first route planning strategy for plannable orders based on the objectives of minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time, and obtains the first delivery route planning decision result for plannable orders.

[0303] Used to determine at least one delivery person to be planned as the target delivery person and at least one order to be planned as the target delivery person, based on the results of the first delivery route planning decision.

[0304] The first path planning module 802 is further specifically as follows:

[0305] Used to generate an initial solution for path planning that satisfies multiple preset path planning constraints, with the initial objective of minimizing the number of delivery personnel involved.

[0306] Used to set the initial solution of path planning as the current solution and the optimal solution of path planning;

[0307] It is used to perform neighborhood search iterations based on the current solution of path planning, and update the current solution and the optimal solution of path planning until the neighborhood search iteration stops;

[0308] This is used to determine the optimal solution for the updated route planning as the first delivery route planning decision result for the plannable orders.

[0309] The first path planning module 802 is further specifically as follows:

[0310] Used to perform neighborhood search based on the current solution of the path planning, to obtain a neighborhood solution of the path planning that satisfies multiple path planning constraints;

[0311] It is used to score the neighborhood solutions of path planning based on multiple preset path planning scoring functions, and obtain the neighborhood solution scoring results;

[0312] Used to update the current solution and the optimal solution of path planning based on the neighborhood solution scoring results;

[0313] Repeat the above steps until the neighborhood search iteration stops.

[0314] The second acquisition module 803 is further specifically as follows:

[0315] Used to obtain real-time delivery route planning trigger commands;

[0316] It is used to obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered based on the real-time delivery route planning trigger instructions.

[0317] The second acquisition module 803 is further specifically as follows:

[0318] Used to determine whether all orders corresponding to a target delivery person have been delivered based on real-time delivery route planning trigger instructions, and

[0319] Used to update the pending orders for the target delivery personnel when not all pending orders have been delivered;

[0320] This is used to terminate the order delivery route planning for the target delivery personnel when all pending orders have been delivered.

[0321] The second path planning module 804 is further defined as follows:

[0322] Used to parse the real-time location address in the real-time location data and the address of the order to be delivered in the order to be delivered, to obtain the latitude and longitude information of the real-time location and the latitude and longitude information of at least one order address to be delivered;

[0323] This is used to perform Cartesian product processing on the union of real-time location latitude and longitude information and at least one order address latitude and longitude information to obtain multiple second flow direction data between real-time location address and at least one order address to be delivered, as well as between different order addresses to be delivered;

[0324] This is used to verify the flow direction of orders to be delivered based on multiple second flow direction data, and to determine the real-time time distance matrix data between orders to be delivered.

[0325] The second path planning module 804 is further defined as follows:

[0326] Used to generate an initial solution for real-time path planning that satisfies multiple preset real-time path planning constraints, based on the objective of minimizing the number of order deliveries.

[0327] Used to set the initial solution of real-time path planning as the current solution and the optimal solution of real-time path planning;

[0328] It is used to perform neighborhood search iterations based on the current solution of real-time path planning, and update the current solution and the optimal solution of real-time path planning until the neighborhood search iteration stops;

[0329] This is used to obtain the optimal solution for real-time route planning after the update is completed, and then use it as the real-time delivery route planning information for the target delivery personnel.

[0330] The second path planning module 804 is further defined as follows:

[0331] Used to perform neighborhood search based on the current solution of real-time path planning, to obtain a neighborhood solution of real-time path planning that satisfies multiple real-time path planning constraints;

[0332] It is used to score the real-time path planning neighborhood solution according to multiple preset real-time path planning scoring functions, and obtain the real-time neighborhood solution scoring result;

[0333] It is used to update the current solution and the optimal solution of real-time path planning based on the real-time neighborhood solution scoring results;

[0334] This is used to repeat the above steps until the neighborhood search iteration stops. In another embodiment of this application, this application also provides a server, such as... Figure 8 As shown, it illustrates a schematic diagram of the server structure involved in an embodiment of this application. Specifically:

[0335] The server may include components such as a processor 901 with one or more processing cores, a memory 902 with one or more computer-readable storage media, a power supply 903, and an input unit 904. Those skilled in the art will understand that... Figure 8 The server structure shown does not constitute a limitation on the server and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. Wherein:

[0336] Processor 901 is the control center of the server, connecting various parts of the server through various interfaces and lines. It performs various functions and processes data by running or executing software programs and / or modules stored in memory 902, and by calling data stored in memory 902, thereby providing overall monitoring of the server. Optionally, processor 901 may include one or more processing cores; processor 901 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. Preferably, processor 901 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the aforementioned modem processor may not be integrated into processor 901.

[0337] The memory 902 can be used to store software programs and modules. The processor 901 executes various functional applications and data processing by running the software programs and modules stored in the memory 902. The memory 902 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the server, etc. In addition, the memory 902 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 902 may also include a memory controller to provide the processor 901 with access to the memory 902.

[0338] The server also includes a power supply 903 that supplies power to the various components. Preferably, the power supply 903 is logically connected to the processor 901 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 903 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0339] The server may also include an input unit 904, which can be used to receive input numeric or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0340] Although not shown, the server may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 901 in the server loads the executable files corresponding to the processes of one or more applications into the memory 902 according to the following instructions, and the processor 901 runs the applications stored in the memory 902 to realize various functions, as follows:

[0341] Obtain the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of delivery personnel to be planned;

[0342] Based on the data from the distribution center, the time-distance matrix data, and the delivery personnel data, and based on the preset first path planning target, the first path planning strategy for the plannable orders is solved to determine at least one delivery personnel to be planned as the target delivery personnel and at least one plannable order as the target delivery personnel for the order to be delivered.

[0343] Obtain real-time location data of the target delivery personnel and real-time time-distance matrix data of the orders to be delivered;

[0344] Based on real-time time-distance matrix data and real-time location data, and with a preset second path planning objective, the second path planning strategy for the order to be delivered is solved to determine the real-time delivery path planning information of the target delivery personnel.

[0345] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0346] In another embodiment of this application, a computer-readable storage medium is also provided, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the order delivery route planning method provided in the embodiments of this application. For example, the computer program loaded by the processor can execute the following steps:

[0347] Obtain the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of delivery personnel to be planned;

[0348] Based on the data from the distribution center, the time-distance matrix data, and the delivery personnel data, and based on the preset first path planning target, the first path planning strategy for the plannable orders is solved to determine at least one delivery personnel to be planned as the target delivery personnel and at least one plannable order as the target delivery personnel for the order to be delivered.

[0349] Obtain real-time location data of the target delivery personnel and real-time time-distance matrix data of the orders to be delivered;

[0350] Based on real-time time-distance matrix data and real-time location data, and with a preset second path planning objective, the second path planning strategy for the order to be delivered is solved to determine the real-time delivery path planning information of the target delivery personnel.

[0351] The above provides a detailed description of an order delivery route planning method, system, server, and storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An order delivery route planning method characterized by, include: The process of acquiring current distribution station data, time-distance matrix data of plannable orders, and delivery personnel data of personnel to be planned includes: acquiring multiple initial orders containing at least one of the plannable orders; parsing the current distribution station address in the distribution station data and the initial order addresses in the multiple initial orders to obtain the latitude and longitude information of the current distribution station and the latitude and longitude information of the multiple initial order addresses; performing Cartesian product processing on the union of the current distribution station latitude and longitude information and the latitude and longitude information of the multiple initial order addresses to obtain multiple first flow direction data between the current distribution station address and the multiple initial order addresses, as well as between different initial order addresses; and performing flow direction verification on the multiple initial orders based on the multiple first flow direction data to determine at least one of the plannable orders among the multiple initial orders and the time-distance matrix data between at least one of the plannable orders. Based on the data from the distribution station, the time-distance matrix data, and the delivery personnel data, and based on a preset first path planning target, the first path planning strategy for the plannable orders is solved to determine at least one of the delivery personnel to be planned as the target delivery personnel and at least one of the plannable orders as the delivery orders of the target delivery personnel. Obtain the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the orders to be delivered; Based on the real-time time distance matrix data and the real-time location data, and based on the preset second path planning target, the second path planning strategy for the order to be delivered is solved to determine the real-time delivery path planning information of the target delivery person.

2. The order delivery route planning method of claim 1, wherein, The first route planning strategy for the plannable orders is solved based on a preset first route planning target to determine at least one of the planned delivery personnel as target delivery personnel and at least one of the plannable orders as orders to be delivered to the target delivery personnel, including: Based on the objectives of minimizing the number of delivery personnel, the average working time of each delivery person, and the order delivery time, the first route planning strategy for the plannable order is solved to obtain the first delivery route planning decision result for the plannable order; Based on the first delivery route planning decision result, at least one of the delivery personnel to be planned is identified as the target delivery personnel, and at least one of the plannable orders is identified as the delivery orders to be delivered by the target delivery personnel.

3. The order delivery route planning method of claim 2, wherein, Solving the first route planning strategy for the plannable orders to obtain the first delivery route planning decision result for the plannable orders includes: With minimizing the number of delivery personnel as the initial objective, an initial solution for path planning that satisfies multiple preset path planning constraints is generated. Set the initial solution of the path planning as the current solution and the optimal solution of the path planning; Based on the current solution of the path planning, a neighborhood search is iterated, and the current solution of the path planning and the optimal solution of the path planning are updated until the neighborhood search iteration stops; The optimal solution of the route planning obtained after the update is used as the first delivery route planning decision result of the plannable order.

4. The order delivery route planning method of claim 3, wherein, The step of performing neighborhood search iterations based on the initial solution of the path planning, and updating the current solution and the optimal solution of the path planning, until the neighborhood search iteration stops, includes: Based on the current solution of the path planning, a neighborhood search is performed to obtain a neighborhood solution of the path planning that satisfies multiple path planning constraints. The neighborhood solution of the path planning is scored according to multiple preset path planning scoring functions to obtain the neighborhood solution scoring result; Based on the neighborhood solution scoring results, the current solution and the optimal solution of the path planning are updated; Repeat the above steps until the neighborhood search iteration stops.

5. The order delivery path planning method of claim 1, wherein, The acquisition of the real-time location data of the target delivery person and the real-time time-distance matrix data of the order to be delivered includes: Obtain real-time delivery route planning trigger commands; Based on the real-time delivery route planning trigger instruction, the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the order to be delivered are obtained.

6. The order delivery route planning method of claim 5, wherein, After obtaining the real-time delivery route planning trigger instruction, the method further includes: Based on the real-time delivery route planning trigger instruction, determine whether all the orders to be delivered corresponding to the target delivery person have been delivered. If not all of the orders to be delivered are completed, the orders to be delivered for the target delivery personnel shall be updated. When all the orders to be delivered are completed, the order delivery route planning for the target delivery personnel ends.

7. The order delivery route planning method of claim 6, wherein, The acquisition of the real-time location data of the target delivery person and the real-time time-distance matrix data of the order to be delivered includes: The real-time location address in the real-time location data and the address of the order to be delivered in the order to be delivered are analyzed to obtain the real-time location latitude and longitude information and at least one address latitude and longitude information of the order to be delivered; The union of the real-time location latitude and longitude information and the latitude and longitude information of at least one of the orders to be delivered is processed by Cartesian product to obtain multiple second flow direction data between the real-time location address and at least one of the orders to be delivered, as well as between different orders to be delivered; Based on multiple second flow direction data, the flow direction of the orders to be delivered is verified to determine the real-time time distance matrix data between the orders to be delivered.

8. The order delivery route planning method of claim 7, wherein, The second route planning strategy for the orders to be delivered is solved based on a preset second route planning objective to determine the real-time delivery route planning information of the target delivery personnel, including: Based on the objective of minimizing the number of order deliveries, an initial solution for real-time path planning that satisfies multiple preset real-time path planning constraints is generated. Set the initial solution of the real-time path planning as the current solution and the optimal solution of the real-time path planning; The neighborhood search is iterated based on the current solution of the real-time path planning, and the current solution and the optimal solution of the real-time path planning are updated until the neighborhood search iteration stops. The optimal solution for real-time route planning obtained after the update is completed will be used as the real-time delivery route planning information for the target delivery personnel.

9. The order delivery path planning method of claim 8, wherein, The step of iterating through neighborhood search based on the current solution of real-time path planning, and updating the current solution and the optimal solution of real-time path planning until the neighborhood search iteration stops, includes: Based on the current solution of the real-time path planning, a neighborhood search is performed to obtain a real-time path planning neighborhood solution that satisfies multiple real-time path planning constraints. The real-time path planning neighborhood solution is scored according to multiple preset real-time path planning scoring functions to obtain the real-time neighborhood solution scoring result. Based on the real-time neighborhood solution scoring results, the current solution and the optimal solution of the real-time path planning are updated. Repeat the above steps until the neighborhood search iteration stops.

10. An order delivery route planning system, characterized in that, The system includes: The first acquisition module is used to acquire the current distribution station data, the time-distance matrix data of plannable orders, and the delivery personnel data of the delivery personnel to be planned. This includes: acquiring multiple initial orders containing at least one of the plannable orders; parsing the current distribution station address in the distribution station data and the initial order addresses in the multiple initial orders to obtain the latitude and longitude information of the current distribution station and the latitude and longitude information of the multiple initial order addresses; performing Cartesian product processing on the union of the current distribution station latitude and longitude information and the latitude and longitude information of the multiple initial order addresses to obtain multiple first flow direction data between the current distribution station address and the multiple initial order addresses, as well as between different initial order addresses; and performing flow direction verification on the multiple initial orders based on the multiple first flow direction data to determine at least one plannable order among the multiple initial orders and the time-distance matrix data between at least one of the plannable orders. The first route planning module is used to solve the first route planning strategy for the plannable orders based on the distribution station data, the time distance matrix data and the delivery personnel data, and a preset first route planning target, to determine at least one of the delivery personnel to be planned as the target delivery personnel and to determine at least one of the plannable orders as the delivery orders of the target delivery personnel. The second acquisition module is used to acquire the real-time location data of the target delivery personnel and the real-time time-distance matrix data of the order to be delivered; The second path planning module is used to solve the second path planning strategy for the order to be delivered based on the real-time time distance matrix data and the real-time location data, and based on the preset second path planning target, to determine the real-time delivery path planning information of the target delivery personnel.

11. A server, characterized in that, The server includes: One or more processors; Memory; and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the order delivery route planning method of any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, It stores a computer program, which is loaded by a processor to execute the steps in the order delivery route planning method according to any one of claims 1 to 9.

Images