Whole vehicle logistics site selection and transportation path planning method and device and electronic equipment

Abstract

The invention provides a whole vehicle logistics site selection and transportation path planning method and device and electronic equipment, and the method comprises the steps: converting the address information of each distribution network point in a preset region range into corresponding geographic position information; cleaning and preprocessing the address information, the geographic position information and the sales volume data of each distribution network to obtain target data, and clustering the target data to obtain a preset number of initial clusters corresponding to a preset area range; and performing iterative optimization on the preset number of initial clusters based on the target data and a preset freight rate data table to obtain a preset number of optimized clusters corresponding to a preset area range, and performing vehicle logistics transportation path planning based on the target data, the optimized clusters and the preset freight rate data table, and obtaining a target whole vehicle logistics transportation path corresponding to the preset area range. By adopting the method, the scientificity and the accuracy of site selection of the whole vehicle logistics transfer warehouse can be improved, and integrated coordination of site selection of the whole vehicle logistics storage center and whole vehicle logistics transportation path planning is realized.

Description

Technical Field

[0001] This invention relates to the field of logistics management technology, and in particular to a method, apparatus and electronic equipment for whole vehicle logistics site selection and transportation route planning. Background Technology

[0002] As the global automotive industry accelerates its transformation towards electrification, intelligentization, and connectivity, automakers' supply chains are facing multiple challenges, including cost reduction and efficiency improvement, low-carbon emission reduction, and flexible response. The digital transformation of automotive logistics has shifted from an "optional" to a "must-have." In the automotive manufacturing industry, vehicle logistics is a crucial link connecting the production and consumption ends, and its efficiency directly impacts operating costs and customer satisfaction. Traditional logistics networks, relying on a single transportation mode (such as highway trunk lines) and a fixed-level warehouse layout, are no longer sufficient to meet the demands of JIT (Just-in-Time) production, regionalized distribution, and unforeseen supply chain risks. Building a smart warehouse network and optimizing multimodal transport routes has become the core direction for automakers' logistics upgrades.

[0003] The site selection for vehicle logistics transit warehouses requires a comprehensive balance of multiple factors, including transportation costs, regional demand density, infrastructure conditions, and policy support. Traditional experience-based decision-making methods are highly subjective and lack the quantifiable analytical ability to analyze the dynamic relationships between these complex variables. Vehicle transportation route planning, as a core component of vehicle logistics, involves various modes of transport, including road, rail, waterway, and multimodal transport. It is characterized by multiple constraints, multiple objectives, and dynamic changes, making traditional manual scheduling or experience-based decision-making methods insufficient to meet current vehicle logistics demands. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method, apparatus and electronic device for whole vehicle logistics site selection and transportation route planning, so as to alleviate the above-mentioned problems existing in the related art.

[0005] In a first aspect, embodiments of the present invention provide a method for whole vehicle logistics site selection and transportation route planning, comprising: converting the address information of each distribution outlet within a preset area into corresponding geographical location information, and cleaning and preprocessing the address information, geographical location information, and sales data of each distribution outlet to obtain target data; clustering the target data to obtain a preset number of initial clusters corresponding to the preset area; wherein each initial cluster includes at least one distribution outlet and each initial cluster has a corresponding initial warehousing center; iteratively optimizing the preset number of initial clusters based on the target data and a preset freight rate data table to obtain a preset number of optimized clusters corresponding to the preset area; and planning whole vehicle logistics transportation routes based on the target data, the optimized clusters, and the preset freight rate data table to obtain the target whole vehicle logistics transportation route corresponding to the preset area.

[0006] Secondly, embodiments of the present invention also provide a vehicle logistics location selection and transportation route planning device, comprising: a data processing module, used to convert the address information of each distribution outlet within a preset area into corresponding geographical location information, and to clean and preprocess the address information, geographical location information, and sales data of each distribution outlet to obtain target data; a clustering module, used to cluster the target data to obtain a preset number of initial clusters corresponding to the preset area; wherein each initial cluster includes at least one distribution outlet and each initial cluster has a corresponding initial warehousing center; an optimization module, used to iteratively optimize the preset number of initial clusters based on the target data and a preset freight rate data table to obtain a preset number of optimized clusters corresponding to the preset area; and a route planning module, used to plan the vehicle logistics transportation route based on the target data, the optimized clusters, and the preset freight rate data table to obtain the target vehicle logistics transportation route corresponding to the preset area.

[0007] Thirdly, embodiments of the present invention also provide an electronic device, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the vehicle logistics site selection and transportation route planning method described in the first aspect above.

[0008] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are invoked and executed by a processor, the computer-executable instructions cause the processor to implement the vehicle logistics location selection and transportation route planning method described in the first aspect above.

[0009] This invention provides a method, apparatus, and electronic device for whole vehicle logistics location selection and transportation route planning. First, the address information of each distribution outlet within a preset area is converted into corresponding geographical location information. Then, the address information, geographical location information, and sales data of each distribution outlet are cleaned and preprocessed to obtain target data. Next, the target data is clustered to obtain a preset number of initial clusters corresponding to the preset area. Then, based on the target data and a preset freight rate table, the preset number of initial clusters are iteratively optimized to obtain a preset number of optimized clusters corresponding to the preset area. Finally, based on the target data, optimized clusters, and the preset freight rate table, whole vehicle logistics transportation route planning is performed to obtain the target whole vehicle logistics transportation route corresponding to the preset area. By employing the aforementioned technologies, address information and sales data of each distribution outlet, along with a pre-set freight rate data table, can be used to achieve intelligent site selection for vehicle logistics warehousing centers through clustering and iterative optimization. This improves the scientific rigor and accuracy of vehicle logistics transit warehouse site selection, optimizes the layout of logistics nodes, and allows for the generation of target vehicle logistics transportation routes within a pre-defined area after the site selection is completed. This enables integrated collaboration between vehicle logistics warehousing center site selection and vehicle logistics transportation route planning, thereby improving the accuracy and efficiency of vehicle logistics operations and promoting the intelligent and digital development of the vehicle logistics transportation network.

[0010] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0011] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0012] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a method for whole vehicle logistics site selection and transportation route planning in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the combination of the K-Means++ algorithm and the EM algorithm in an embodiment of the present invention. Figure 3This is a flowchart of the Dijikstra algorithm in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a vehicle logistics site selection and transportation route planning device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. 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.

[0015] Currently, existing technologies for selecting transshipment warehouses for vehicle logistics rely heavily on experience-based decision-making, which is highly subjective and makes it difficult to quantify the dynamic relationships between these complex variables. Vehicle transportation route planning involves multiple modes of transportation and has characteristics such as multiple constraints, multiple objectives, and dynamic changes. Traditional manual scheduling or experience-based decision-making methods are no longer sufficient to meet the current needs of vehicle logistics transportation.

[0016] Based on this, the present invention provides a method, apparatus and electronic device for whole vehicle logistics site selection and transportation route planning, which can alleviate the above-mentioned problems existing in related technologies.

[0017] To facilitate understanding of this embodiment, a detailed description of the whole vehicle logistics site selection and transportation route planning method disclosed in this embodiment of the invention will be provided first. (See [link to relevant documentation]). Figure 1 As shown, the method may include the following steps: Step S102: Convert the address information of each distribution outlet within the preset area into corresponding geographical location information, and clean and preprocess the address information, geographical location information and sales data of each distribution outlet to obtain the target data.

[0018] During the data collection phase, it is necessary to comprehensively obtain relevant information about all dealerships of the car manufacturer. This information mainly includes: dealership address information, dealership sales data, and other relevant data.

[0019] The specific method for obtaining distribution outlet address information is as follows: collect the detailed addresses of all distribution outlets as distribution outlet address information. This data is usually obtained from the company's internal database.

[0020] The specific method for obtaining sales data from distribution outlets can be as follows: collect the annual historical sales data of each distribution outlet as the sales data of the distribution outlet. The sales data of the distribution outlet can help assess the importance and demand scale of each distribution outlet, and provide a basis for weighting in subsequent cluster analysis.

[0021] Other relevant data may specifically refer to the inventory capacity, transportation costs, and service scope of the distribution network. Other relevant data can be used to further optimize the results obtained after subsequent cluster analysis.

[0022] Based on the address information of the distribution outlets, the geographical location latitude and longitude of all distribution outlets can be obtained. Specifically, geocoding tools (such as APIs of specific map services) can be used to convert the detailed addresses of the distribution outlets into latitude and longitude coordinates in batches for geospatial analysis.

[0023] After data collection is complete, the data needs to be cleaned and preprocessed to ensure its accuracy and consistency. This includes checking for correct addresses, complete sales data, and accurate latitude and longitude coordinates.

[0024] Step S104: Cluster the target data to obtain a preset number of initial clusters corresponding to a preset area range.

[0025] Each initial cluster may include at least one distribution outlet, and each initial cluster may have a corresponding initial warehousing center.

[0026] Step S106: Based on the target data and the preset fare data table, iteratively optimize the preset number of initial clusters to obtain the preset number of optimized clusters corresponding to the preset area range.

[0027] Step S108: Based on the target data, the optimized cluster, and the preset freight rate data table, plan the whole vehicle logistics transportation route to obtain the target whole vehicle logistics transportation route corresponding to the preset area range.

[0028] This invention provides a method for whole vehicle logistics location selection and transportation route planning. First, the address information of each distribution outlet within a preset area is converted into corresponding geographical location information. Then, the address information, geographical location information, and sales data of each distribution outlet are cleaned and preprocessed to obtain target data. Next, the target data is clustered to obtain a preset number of initial clusters corresponding to the preset area. Then, based on the target data and a preset freight rate data table, the preset number of initial clusters are iteratively optimized to obtain a preset number of optimized clusters corresponding to the preset area. Finally, based on the target data, optimized clusters, and preset freight rate data table, whole vehicle logistics transportation route planning is performed to obtain the target whole vehicle logistics transportation route corresponding to the preset area. By employing the aforementioned technologies, address information and sales data of each distribution outlet, along with a pre-set freight rate data table, can be used to achieve intelligent site selection for vehicle logistics warehousing centers through clustering and iterative optimization. This improves the scientific rigor and accuracy of vehicle logistics transit warehouse site selection, optimizes the layout of logistics nodes, and allows for the generation of target vehicle logistics transportation routes within a pre-defined area after the site selection is completed. This enables integrated collaboration between vehicle logistics warehousing center site selection and vehicle logistics transportation route planning, thereby improving the accuracy and efficiency of vehicle logistics operations and promoting the intelligent and digital development of the vehicle logistics transportation network.

[0029] As one possible implementation, the target data may include the first location coordinates and sales volume of each distribution outlet; based on this, step S104 (i.e., clustering the target data to obtain a preset number of initial clusters corresponding to a preset area range) may include: Step A1: Set the corresponding vehicle demand for each dealership based on its sales volume.

[0030] Following the previous example, after obtaining the annual historical sales data and latitude and longitude coordinates of all dealerships, the annual historical sales data can be used to assess the importance and demand scale of each dealership, and then the corresponding vehicle demand volume can be set as weight information for each dealership based on its importance and demand scale.

[0031] Step A2: Based on the total vehicle demand and first location coordinates of all dealerships, a preset clustering algorithm is used to cluster all dealerships into a preset number of initial clusters.

[0032] The preset clustering algorithm can be K-Means clustering algorithm, mean shift clustering algorithm, DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, hierarchical clustering algorithm, etc., and there is no limitation on it.

[0033] For example, each initial storage center may have its own corresponding second location coordinates; the first initial storage center may be randomly generated based on the total vehicle demand of all distribution outlets and the first location coordinates; then, iterative calculations are performed based on the total vehicle demand of each distribution outlet, the first location coordinates, and the second location coordinates of the first initial storage center to generate each subsequent initial storage center until the total number of all initial storage centers reaches a preset number, at which point the generation of initial storage centers stops; wherein, after each initial storage center is generated, the target weighted distance between each distribution outlet and its corresponding target initial storage center with the smallest distance is calculated based on the total vehicle demand of each distribution outlet, the first location coordinates, and the second location coordinates of each current initial storage center, and the first location coordinates of each distribution outlet and the obtained target weighted distance is used to generate the next initial storage center.

[0034] In practical applications, the steps described above for calculating the target weighted distance between each dealership and its corresponding target initial warehouse center, based on the vehicle demand volume and first location coordinates of each dealership and the second location coordinates of each current initial warehouse center, may include: determining the target initial warehouse center corresponding to each dealership based on the first location coordinates of each dealership and the second location coordinates of each current initial warehouse center; and for each dealership, calculating the target weighted distance between the dealership and its corresponding target initial warehouse center based on the vehicle demand volume and first location coordinates of that dealership and the second location coordinates of the target initial warehouse center corresponding to that dealership.

[0035] Accordingly, the steps of generating the next initial storage center based on the first location coordinates of each distribution point and the obtained target weighted distance may include: calculating the probability corresponding to each distribution point based on the obtained target weighted distance, and generating the next initial storage center based on the first location coordinates of each distribution point and the obtained probability.

[0036] For ease of understanding, the clustering method of the target data is described below using the K-Means++ algorithm as an example of the preset clustering algorithm.

[0037] The steps for clustering using the K-Means++ algorithm are as follows: 1) Data preparation: Collect relevant data on vehicle logistics, including the location coordinates (such as latitude and longitude coordinates) of all dealer outlets. Each dealer outlet The total demand for vehicles (i.e., the weight). Each dealer outlet The transportation costs and the number of warehousing centers (K) to be located.

[0038] 2) Initialize cluster centers: Use the K-Means++ algorithm to randomly select initial cluster centers from the dealer network.

[0039] The specific steps for selecting initial cluster centers may include the following steps 21) to 22): 21) Randomly select the first center: Randomly select the first warehousing center based on the probability distribution of vehicle demand at dealer outlets. , These are the position coordinates; 22) Iterative selection of subsequent warehouse centers: When selecting a warehouse center in each iteration, based on... , And the selected fulfillment centers (only the first fulfillment center is included if a second fulfillment center needs to be selected if only the first one is selected). Calculate each dealer outlet With the selected nearest center (i.e., with the dealer network) Weighted distance of the warehouse center with the smallest distance between them , For dealer outlets The minimum distance to all distribution centers (i.e., the distance to each distributor outlet) (distance value from the selected nearest center), based on probability Select the next storage center Probability of repeated execution And based on probability Select the next storage center The process continues until there are K selected warehouse centers (i.e., K warehouse centers have been selected), at which point the warehouse center selection is complete.

[0040] As one possible implementation, the aforementioned preset fare data table may include the unit price of different transportation modes; based on this, the aforementioned step S106 (i.e., iteratively optimizing the preset number of initial clusters based on the target data and the preset fare data table to obtain the preset number of optimized clusters corresponding to the preset area range) may include: using the EM algorithm to iteratively optimize the preset number of initial clusters based on the target data and the preset fare data table to obtain the preset number of optimized clusters.

[0041] For example, each optimized cluster may have a corresponding optimized storage center; the weighted distance between each distribution point and each initial storage center can be calculated based on the first location coordinates and vehicle demand of each distribution point and the second location coordinates of each initial storage center; and the initial local transportation cost from each distribution point to each initial storage center can be calculated based on the transportation unit price of the target transportation mode corresponding to each distribution point and the weighted distance between each distribution point and each initial storage center; then the initial transportation cost is calculated based on the obtained initial local transportation cost; then, with the goal of minimizing the transportation cost, the following operations (a1) to (a3) ​​are performed iteratively on a preset number of initial clusters based on the initial transportation cost: (a1) For each distribution point, based on the first location coordinates of the distribution point and the demand for the whole vehicle and the second location coordinates of each current warehouse center, calculate the weighted distance between the distribution point and each current warehouse center, and calculate the membership probability of the distribution point to each current warehouse center based on the weighted distance obtained for the distribution point. (a2) Based on the first location coordinates and vehicle demand of all target dealerships with a membership probability greater than a preset probability threshold corresponding to the same current warehouse center, and the membership probability of all target dealerships to the same current warehouse center, calculate the weighted geographic center of all target dealerships corresponding to the same current warehouse center, and move the same current warehouse center to the location of its corresponding weighted geographic center; (a3) Based on the unit price of the target transportation mode corresponding to each distribution outlet and the weighted distance between each distribution outlet and each current storage center, calculate the current local transportation cost from each distribution outlet to each current storage center, and update the current transportation cost based on the obtained current local transportation cost; wherein, the current transportation cost is the initial transportation cost for the first iteration; The above operations (a1) to (a3) ​​are performed iteratively until the preset maximum number of iterations is reached and the iteration stops. The preset number of current storage centers obtained in the last iteration is taken as the preset number of optimized storage centers. All target distribution outlets with a membership probability greater than a preset probability threshold corresponding to the same optimized storage center are grouped into corresponding optimized clusters to obtain the preset number of optimized clusters.

[0042] To facilitate understanding, the operation method of iterative optimization using the EM algorithm is described below using a specific application as an example.

[0043] The steps for iterative optimization using the EM algorithm are as follows: (1) Initialization: Set the count value of the iteration counter (representing the current iteration number). Set the maximum number of iterations and the initial transportation cost. The initialization results obtained after clustering the target data using the K-Means++ algorithm are obtained. .

[0044] (2) Loop: The loop is mainly divided into two parts: the E step and the M step.

[0045] Step E: For each customer point (i.e., dealer outlet) ), calculate its weighted distance to each distribution center, and then use the weighted distance to each distributor outlet. Calculate its membership probability to each storage center. , and Distributor outlets to the warehousing center and The weighted distance, and Distributor outlets to the warehousing center and distance, These are the coefficients used to calculate the membership probability.

[0046] M-step: For each warehouse center, collect all dealer outlets (i.e., target dealer outlets, or target customer points) that have a membership relationship with it (e.g., the probability of membership is greater than a certain probability threshold). Considering the vehicle demand and membership strength (i.e., the probability of membership) of each target customer point, calculate the weighted geographical center of these target customer points corresponding to each warehouse center. Move each warehouse center to the location of the weighted geographical center of its corresponding target customer point and update the status (including all current warehouse centers obtained after this iteration). ), update the iteration counter value (representing the current iteration number). Calculate the transportation cost after this iteration. The process continues until the maximum number of iterations is reached, at which point the iteration stops. The final result (including the location coordinates of all warehouse centers obtained after the last iteration and the corresponding clusters composed of the target customer points corresponding to each warehouse center) is used as the warehouse center location result.

[0047] As one possible implementation, step S108 (i.e., planning the full-vehicle logistics transportation route based on the target data, the optimized cluster, and the preset freight rate data table to obtain the target full-vehicle logistics transportation route corresponding to the preset area range) may include: Step B1: Based on the first location coordinates and vehicle demand of each distribution outlet and the second location coordinates of each optimized warehousing center, calculate the weighted distance between each distribution outlet and each optimized warehousing center. Based on the transportation unit price of the target transportation mode corresponding to each distribution outlet and the weighted distance between each distribution outlet and each optimized warehousing center, calculate the local transportation cost from each distribution outlet to each optimized warehousing center.

[0048] Step B2: Based on the first location coordinates of each distribution outlet, the second location coordinates of each optimized warehousing center, the third location coordinates of the base warehouse, and the obtained local transportation costs, construct a transportation network corresponding to the preset area.

[0049] In this transportation network, nodes represent base warehouses, distribution outlets, and optimized storage centers; edges between nodes represent feasible routes for the target transportation mode; and the weight of the edges represents local transportation costs.

[0050] Step B3: Use a preset shortest path algorithm to plan the shortest path for the transportation network, and use the obtained shortest path as the target vehicle logistics transportation route.

[0051] The aforementioned preset shortest path algorithm can specifically adopt Dijkstra's algorithm, Floyd's algorithm, Bellman-Ford algorithm, SPFA algorithm, etc., without limitation.

[0052] Taking the Dijkstra algorithm as an example of the preset shortest path algorithm, the above whole vehicle logistics transportation route planning method is as follows: b1) Construct a transportation network that includes base warehouses, transit warehouses (i.e., warehousing centers), and sales outlets (i.e., distribution outlets). In the transportation network, nodes represent logistics nodes (base warehouses, transit warehouses, and sales outlets), and edges represent feasible routes for different modes of transportation (road, rail, and waterway). The weight of the edges in the transportation network is determined by the local transportation cost between nodes (calculated based on the real-time quotations of road, rail, and waterway transportation, the demand for vehicles at each node, and the distance between nodes).

[0053] b2) Initialization: b21) Create distance array , Used to record data from the initial starting point (i.e., the base repository). The current known lowest cost value to each node (i.e., the minimum local transportation cost from the initial starting point to all nodes) is used to determine the initial starting point. Set the current known lowest cost value in the table to 0, and set all other nodes in the table to 0. The current known lowest cost value is initialized to infinity (indicating that the cost is unknown or unattainable).

[0054] b22) Create a predecessor node array , This is used to record the previous node on the optimal path to each node, creating an array of predecessor transportation methods. , This is used to record the transportation method used to reach each node, creating a priority queue. Add all nodes And sort them in ascending order of the currently known lowest cost value.

[0055] b3) Main loop: Repeat the following steps until the priority queue is reached. Empty: b31) Select the node with the smallest known lowest cost value from the priority queue. Remove from and take out of the middle The node with the smallest known minimum cost value is denoted as... ; b32) Traverse the neighbors and perform a "relaxation" operation: for node Each neighbor node (i.e., there exists from) point to (the edge) performs the following operations: b321) Calculate candidate costs: Calculate node Each neighbor node The candidate cost is , From the initial starting point to the node The current known lowest cost value, function According to point to edge Search for the corresponding freight rate or fixed fee based on the type of transport (road transport / rail transport / water transport) and return the results. Local transportation costs.

[0056] b322) Comparison and Update: If Less than from the initial starting point to the node The current known lowest cost value Then update make ,renew Make the node reachable The previous node on the optimal path , indicating arrival at node The best path is from node ,renew The mode of transportation used to reach the node For the edge Type, update priority queue Based on nodes Reduced Readjust nodes In the priority queue The sorting position in the queue ensures priority. The nodes in the system are always sorted in ascending order of the currently known lowest cost value.

[0057] B4) Termination and Path Reconstruction: When the priority queue When the value is empty, Dijkstra's algorithm ends. The data stored in it is from the initial starting point. Find the lowest cost to all nodes in the transportation network, and label the transportation method corresponding to each edge on the path formed at this lowest cost.

[0058] To facilitate understanding, the implementation of the above-mentioned whole vehicle logistics location selection and transportation route planning method is described below using a specific application as an example.

[0059] To implement the aforementioned vehicle logistics location selection and transportation route planning method, the K-Means++ algorithm can be introduced for intelligent initialization of warehouse center locations to avoid the EM algorithm getting trapped in local optima. Then, the EM algorithm is introduced to iteratively optimize the initial warehouse center location scheme obtained by the K-Means++ algorithm, continuously improving the warehouse center location through E-steps and M-steps. In other words, the aforementioned vehicle logistics location selection and transportation route planning method can actually provide a vehicle logistics location scheme for automakers based on a combination of the K-Means++ and EM algorithms, minimizing the total transportation cost from all dealer outlets to their nearest warehouse center. Furthermore, based on the vehicle logistics location selection, the aforementioned method can further introduce the Dijkstra algorithm for optimal route planning, effectively providing an optimal route planning scheme for multimodal vehicle logistics transportation based on the Dijkstra algorithm.

[0060] The implementation methods of the above-mentioned whole vehicle logistics site selection and transportation route planning methods can mainly include: (a) Data collection and processing.

[0061] It can comprehensively collect relevant information from all dealerships of car manufacturers (including dealership address information, dealership sales data, and other relevant data), and convert dealership address information into geographical location latitude and longitude information (including the latitude and longitude coordinates of each dealership). The obtained data (including relevant information and geographical location latitude and longitude information) is cleaned and preprocessed, such as checking whether the address is correct, whether the sales data is complete, and whether the latitude and longitude are accurate, to ensure the accuracy and consistency of the data.

[0062] (ii) Using the K-Means++ algorithm combined with the EM algorithm for vehicle logistics site selection for car manufacturers.

[0063] The computation process of the K-Means++ algorithm combined with the EM algorithm can be mainly divided into an initialization phase and an iterative optimization phase.

[0064] Initialization phase: The K-Means++ clustering algorithm was written using Python to complete... Figure 2 The K-Means++ clustering algorithm part of the calculation process is shown in the figure, in order to achieve... Figure 2 The calculation process of the K-Means++ clustering algorithm yields the initial warehouse center location results and the range of distributors covered by the shortest distance target; Iterative optimization phase: Introducing the EM algorithm to complete... Figure 2 The EM algorithm section of the calculation process is shown in the figure, in order to... Figure 2 The calculation process of the EM algorithm combines the initial warehouse location results and the range of distributors it covers, as well as the freight rates of road, rail and water transportation, to calculate the new warehouse location results and the range of distributors it covers under the weighted distance cost of all distributors.

[0065] See Figure 2 As shown, the process of combining the K-Means++ algorithm with the EM algorithm is as follows: Step 1, Initialization, setting parameters (iteration counter count value) Maximum number of iterations ); Step 2: Initialize the results using the K-Means++ algorithm (including obtaining the initial warehouse center location through clustering and calculating the initial total cost).

[0066] Step 3: Use the EM algorithm for iterative optimization.

[0067] The loop begins by checking if the stopping condition is met (i.e., the maximum number of iterations has been reached). If the stopping condition is not met, the iteration continues. If the stopping condition is met, the location of the warehouse center is determined and the final result (including the optimal warehouse center location and its distributor affiliation) is output.

[0068] The iterative calculation process mainly includes: E-step (calculating membership probabilities), M-step (updating warehouse center location), updating the state (including warehouse center location), and updating the iteration counter value. Calculate the transportation cost after this iteration, then proceed to the next iteration, returning to the beginning of the loop to check if the stopping condition is met.

[0069] The specific calculation process of membership probability: For each distributor point Calculate it to each warehouse center Weighted distance Calculate each dealer outlet based on weighted distance probability of membership degree of each warehousing center .

[0070] The specific calculation process for updating the location of a storage center: For each storage center Based on the membership probability, all affiliated dealers are collected, and a weighted geographic center is calculated considering the size of the vehicle demand and the membership probability. The warehousing center is then moved to the weighted geographic center.

[0071] (III) Optimal route planning for multimodal transport logistics based on Dijkstra's algorithm.

[0072] The optimal route planning for multimodal transport logistics based on Dijkstra's algorithm can be mainly divided into three stages: data collection, optimal route solution, and optimal route output.

[0073] Data collection phase: Obtain the location of the base warehouse, transit warehouse, dealer location, demand of all dealers (i.e., demand for complete vehicles), and freight rate data tables for all regions (including freight rate data for road transport, rail transport, and water transport).

[0074] Optimal path finding stage: Substitute all the data collected in the data collection stage into the Dijikstra algorithm to find the optimal path.

[0075] Optimal Path Output Stage: Outputs the optimal path (including the starting point, transfer point, destination, and transportation methods between each node) obtained in the optimal path solution stage.

[0076] See Figure 3 As shown, the process of Dijkstra's algorithm is as follows: Step 1, Initialization: Creation , , , Set the initial starting point at The value in (which can be simplified to) The value of ) is 0 while the values ​​of the other nodes are 0. The value is infinity, According to the middle node Sort the values ​​in ascending order.

[0077] Step 2, Main Loop: Judgment Is it empty; when When not empty, from Take out The node with the smallest value traversal Neighbors Calculate candidate costs ,like Less than Then update , , ,Adjustment middle The sorting position continues from Extract the next node (i.e.) middle (find the node with the smallest value) and return to the judgment. Is the step empty? If Not less than Then return to the judgment. The step to determine if the value is empty.

[0078] Step 3, End and Optimal Path Reconstruction: Until The process ends when the value is empty; output the result. , , Reconstruct the optimal path.

[0079] The beneficial effects of the above-mentioned whole vehicle logistics site selection and transportation route planning methods are mainly reflected in the following aspects: Using the combination of the K-Means++ algorithm and the EM algorithm for the site selection of whole vehicle logistics warehousing centers improves the scientificity and accuracy of whole vehicle logistics transit warehouse site selection, achieving optimal logistics node layout; Utilizing the Dijkstra algorithm to optimize whole vehicle transportation route planning under multimodal transport achieves the comprehensive optimization of transportation costs and time, improving operational accuracy and efficiency; Realizing integrated collaborative design of whole vehicle logistics transit warehouse site selection and whole vehicle transportation route planning promotes the intelligent and digital development of whole vehicle logistics networks.

[0080] Based on the above-described method for vehicle logistics location selection and transportation route planning, this invention also provides a device for vehicle logistics location selection and transportation route planning. (See attached image) Figure 4 As shown, the device may include the following modules: The data processing module 402 is used to convert the address information of each distribution outlet within the preset area into corresponding geographical location information, and to clean and preprocess the address information, geographical location information and sales data of each distribution outlet to obtain target data. Clustering module 404 is used to cluster the target data to obtain a preset number of initial clusters corresponding to a preset area range; wherein each initial cluster includes at least one distribution outlet and each initial cluster has a corresponding initial storage center. Optimization module 406 is used to iteratively optimize a preset number of initial clusters based on the target data and a preset fare data table to obtain a preset number of optimized clusters corresponding to the preset area range; The route planning module 408 is used to plan the whole vehicle logistics transportation route based on the target data, the optimized cluster and the preset freight rate data table, so as to obtain the target whole vehicle logistics transportation route corresponding to the preset area range.

[0081] By employing the aforementioned vehicle logistics location selection and transportation route planning device, the address information and sales data of each distribution outlet, as well as a preset freight rate data table, can be used to achieve intelligent location selection of vehicle logistics warehousing centers through clustering and iterative optimization. This improves the scientificity and accuracy of vehicle logistics transit warehouse location selection, optimizes the layout of logistics nodes, and allows for the generation of target vehicle logistics transportation routes corresponding to preset areas after the location selection of the vehicle logistics warehousing center is completed. This achieves integrated collaboration between vehicle logistics warehousing center location selection and vehicle logistics transportation route planning, which helps improve the accuracy and efficiency of vehicle logistics operations, thereby promoting the intelligent and digital development of the vehicle logistics transportation network.

[0082] The vehicle logistics location selection and transportation route planning device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned vehicle logistics location selection and transportation route planning method embodiment. For the sake of brevity, any parts not mentioned in the vehicle logistics location selection and transportation route planning device embodiment can be referred to the corresponding content in the aforementioned vehicle logistics location selection and transportation route planning method embodiment.

[0083] This invention also provides an electronic device, such as... Figure 5 The diagram shows the structure of the electronic device 100, which includes a processor 51 and a memory 50. The memory 50 stores computer-executable instructions that can be executed by the processor 51. The processor 51 executes the computer-executable instructions to implement the above-mentioned whole vehicle logistics location selection and transportation route planning method.

[0084] exist Figure 5In the illustrated embodiment, the electronic device 100 further includes a bus 52 and a communication interface 53, wherein the processor 51, the communication interface 53, and the memory 50 are connected via the bus 52.

[0085] The memory 50 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 53 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 52 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 52 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0086] The processor 51 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above-mentioned vehicle logistics location selection and transportation route planning method can be completed by the integrated logic circuits in the processor 51 or by software instructions. The processor 51 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the vehicle logistics location selection and transportation route planning method disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory. The processor 51 reads the information in the memory and, in conjunction with its hardware, completes the steps of the vehicle logistics location selection and transportation route planning method of the aforementioned embodiment.

[0087] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the above-described whole vehicle logistics location selection and transportation route planning method. For specific implementation details, please refer to the foregoing method embodiments, which will not be repeated here.

[0088] The computer program product of the whole vehicle logistics location selection and transportation route planning method, device and electronic device provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0089] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0090] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0092] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for site selection and transportation route planning in vehicle logistics, characterized in that, include: The address information of each distribution outlet within the preset area is converted into corresponding geographic location information, and the address information, geographic location information and sales data of each distribution outlet are cleaned and preprocessed to obtain the target data. The target data is clustered to obtain a preset number of initial clusters corresponding to a preset area range; wherein each initial cluster includes at least one distribution outlet and each initial cluster has a corresponding initial storage center; Based on the target data and the preset fare data table, the preset number of initial clusters are iteratively optimized to obtain the preset number of optimized clusters corresponding to the preset area range; Based on the target data, the optimized cluster, and the preset freight rate data table, a full vehicle logistics transportation route is planned to obtain the target full vehicle logistics transportation route corresponding to the preset area range.

2. The method for whole vehicle logistics site selection and transportation route planning according to claim 1, characterized in that, The target data includes the first location coordinates and sales volume of each distribution outlet; Clustering the target data to obtain a preset number of initial clusters corresponding to a preset region range includes: Set a corresponding vehicle demand for each dealership based on its sales volume. Based on the total vehicle demand and first location coordinates of all dealerships, a pre-defined clustering algorithm is used to cluster all dealerships into a pre-defined number of initial clusters.

3. The method for whole vehicle logistics site selection and transportation route planning according to claim 2, characterized in that, Each initial storage center has its own corresponding second location coordinates; Based on the total vehicle demand and initial location coordinates of all dealerships, a pre-defined clustering algorithm is used to cluster all dealerships into a predetermined number of initial clusters, including: Based on the vehicle demand and first location coordinates of all dealerships, the first initial warehouse center is randomly generated. Based on the vehicle demand and first location coordinates of each dealership, and the second location coordinates of the first initial warehouse center, iterative calculations are performed to generate each subsequent initial warehouse center until the total number of initial warehouse centers reaches a preset number, at which point the generation of initial warehouse centers stops. Specifically, after each initial warehouse center is generated, based on the vehicle demand and first location coordinates of each dealership, and the second location coordinates of each current initial warehouse center, the target weighted distance between each dealership and its corresponding target initial warehouse center with the smallest distance is calculated. Then, based on the first location coordinates of each dealership and the obtained target weighted distance, the next initial warehouse center is generated.

4. The method for whole vehicle logistics site selection and transportation route planning according to claim 3, characterized in that, Based on the vehicle demand and first location coordinates of each dealership, and the second location coordinates of each current initial warehouse center, calculate the target weighted distance between each dealership and its corresponding target initial warehouse center with the smallest distance. This includes: determining the target initial warehouse center corresponding to each dealership based on the first location coordinates of each dealership and the second location coordinates of each current initial warehouse center; and for each dealership, calculating the target weighted distance between the dealership and its corresponding target initial warehouse center based on the vehicle demand and first location coordinates of that dealership and the second location coordinates of its corresponding target initial warehouse center. The process involves generating the next initial storage center based on the first location coordinates of each distribution point and the obtained target weighted distance, including: calculating the probability corresponding to each distribution point based on the obtained target weighted distance, and generating the next initial storage center based on the first location coordinates of each distribution point and the obtained probability.

5. The method for whole vehicle logistics site selection and transportation route planning according to claim 2, characterized in that, The preset freight rate data table includes the unit price of different transportation modes; Based on the target data and the preset fare data table, the preset number of initial clusters are iteratively optimized to obtain the preset number of optimized clusters corresponding to the preset area range, including: Based on the target data and the preset fare data table, the EM algorithm is used to iteratively optimize a preset number of initial clusters to obtain a preset number of optimized clusters.

6. The method for whole vehicle logistics site selection and transportation route planning according to claim 5, characterized in that, Each optimized cluster has its own corresponding optimized storage center; based on the target data and the preset freight rate data table, the EM algorithm is used to iteratively optimize a preset number of initial clusters to obtain a preset number of optimized clusters, including: Based on the first location coordinates and vehicle demand of each distribution outlet and the second location coordinates of each initial storage center, the weighted distance between each distribution outlet and each initial storage center is calculated. Based on the transportation unit price of the target transportation mode corresponding to each distribution outlet and the weighted distance between each distribution outlet and each initial storage center, the initial local transportation cost from each distribution outlet to each initial storage center is calculated. Then, the initial transportation cost is calculated based on the obtained initial local transportation cost. With the goal of minimizing transportation costs, the following operations are performed on a preset number of initial clusters based on the initial transportation costs: For each distribution point, based on the first location coordinates of the distribution point and the demand for vehicles, as well as the second location coordinates of each current storage center, the weighted distance between the distribution point and each current storage center is calculated, and the membership probability of the distribution point to each current storage center is calculated based on the weighted distance obtained for the distribution point. Based on the first location coordinates and vehicle demand of all target dealerships with a membership probability greater than a preset probability threshold corresponding to the same current warehouse center, as well as the membership probability of all target dealerships to the same current warehouse center, calculate the weighted geographic center of all target dealerships corresponding to the same current warehouse center, and move the same current warehouse center to the location of its corresponding weighted geographic center. Based on the unit price of the target transportation mode corresponding to each distribution outlet and the weighted distance between each distribution outlet and each current storage center, the current local transportation cost from each distribution outlet to each current storage center is calculated, and the current transportation cost is updated based on the obtained current local transportation cost; wherein, the current transportation cost for the first iteration is the initial transportation cost; The iteration stops when the preset maximum number of iterations is reached. The preset number of current storage centers obtained in the last iteration is taken as the preset number of optimized storage centers. All target distribution outlets with a membership probability greater than a preset probability threshold corresponding to the same optimized storage center are grouped into corresponding optimized clusters to obtain the preset number of optimized clusters.

7. The method for whole vehicle logistics site selection and transportation route planning according to claim 6, characterized in that, Based on the target data, the optimized cluster, and the preset freight rate data table, a full vehicle logistics transportation route is planned to obtain the target full vehicle logistics transportation route corresponding to the preset area, including: Based on the first location coordinates and vehicle demand of each distribution outlet and the second location coordinates of each optimized warehouse center, the weighted distance between each distribution outlet and each optimized warehouse center is calculated. Based on the transportation unit price of the target transportation mode corresponding to each distribution outlet and the weighted distance between each distribution outlet and each optimized warehouse center, the local transportation cost from each distribution outlet to each optimized warehouse center is calculated. Based on the first location coordinates of each distribution outlet, the second location coordinates of each optimized warehousing center, and the third location coordinates of the base warehouse, as well as the obtained local transportation costs, a transportation network corresponding to a preset area is constructed. In the transportation network, nodes represent the base warehouse, distribution outlets, and optimized warehousing centers, edges between nodes represent feasible routes of the target transportation mode, and the weight of the edges represents the local transportation costs. A preset shortest path algorithm is used to plan the shortest path for the transportation network, and the obtained shortest path is used as the target whole vehicle logistics transportation path.

8. A device for whole vehicle logistics site selection and transportation route planning, characterized in that, include: The data processing module is used to convert the address information of each distribution outlet within the preset area into corresponding geographical location information, and to clean and preprocess the address information, geographical location information and sales data of each distribution outlet to obtain the target data. The clustering module is used to cluster the target data to obtain a preset number of initial clusters corresponding to a preset area range; wherein each initial cluster includes at least one distribution outlet and each initial cluster has a corresponding initial storage center. The optimization module is used to iteratively optimize a preset number of initial clusters based on the target data and a preset fare data table to obtain a preset number of optimized clusters corresponding to the preset area range. The route planning module is used to plan the whole vehicle logistics transportation route based on the target data, the optimized cluster and the preset freight rate data table, so as to obtain the target whole vehicle logistics transportation route corresponding to the preset area range.

9. An electronic device, characterized in that, The system includes a processor and a memory, the memory storing computer-executable instructions that can be executed by the processor, the processor executing the computer-executable instructions to implement the whole vehicle logistics location selection and transportation route planning method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the whole vehicle logistics location selection and transportation route planning method according to any one of claims 1 to 7.

Images