Intelligent planning method and system for tobacco logistics under high-density distribution constraints

Abstract

The present invention discloses a method and system for intelligent planning of tobacco logistics under high-density distribution condition constraints, the method comprising: clustering and dividing the geographical locations of retailers to be distributed to form multiple business districts; taking the business district as the basic unit, forming multiple single-vehicle distribution business district sets according to the high-density distribution constraint conditions, or generating a collection area route containing all business districts according to a dynamic route optimization algorithm; and planning the distribution paths of each transport vehicle according to multiple single-vehicle distribution business district sets or collection area routes. The present invention can make the workload of each vehicle uniform, the loading rate high, and the distribution efficiency greatly improved under the condition of a large number of constraint conditions applicable to actual situations.

Description

Technical Field

[0001] The present invention belongs to the technical field of tobacco logistics planning, and in particular relates to a tobacco logistics intelligent planning method and system under the constraints of high-density distribution conditions. Background Art

[0002] In recent years, the logistics field has developed very rapidly, and there have been many studies related to logistics. However, in the field of tobacco logistics and distribution, due to the distribution characteristics of distribution outlets that are often highly concentrated in local areas of cities, coupled with the management regulations of tobacco monopoly, tobacco companies face technical difficulties such as high costs and difficult path optimization when independently building distribution networks. The distribution volume of tobacco logistics is large, the distribution volume varies greatly periodically, and the distribution outlets cover a large area, but the distribution volume is often concentrated in a few areas. This has led to many existing cutting-edge technologies in the field of logistics not being well applied to the field of tobacco logistics.

[0003] In the field of tobacco logistics, transport vehicles are the main logistics costs and resources of tobacco companies. So how to maximize the utilization of distribution resources and achieve more efficient and reasonable vehicle loading can greatly reduce tobacco distribution costs and realize the modernization of tobacco logistics.

[0004] Existing vehicle scheduling algorithms for tobacco logistics often only focus on individual issues such as clustering of intermediate distribution stations or full load rates, but fail to propose a complete system model and solution. In addition, there are basically no experiments with real data in a real tobacco distribution environment, which makes it difficult to implement in actual applications or has poor results. Summary of the invention

[0005] In view of this, the present invention proposes a tobacco logistics intelligent planning method and system under the constraints of high-density distribution conditions, which are used to solve the problem of low loading rate of transport vehicles under high-density logistics distribution conditions.

[0006] In a first aspect, the present invention discloses a method for intelligent planning of tobacco logistics under high-density distribution constraints, the method comprising:

[0007] Cluster the geographical locations of retailers to be delivered to form multiple business districts;

[0008] Taking the business district as the basic unit, multiple single-bike delivery business district sets are formed according to the constraints of high-density delivery, or a collection area route including all business districts is generated according to the dynamic route optimization algorithm;

[0009] The delivery routes for each transport vehicle are planned based on multiple bicycle delivery business district collections or area lines.

[0010] On the basis of the above technical solution, preferably, the forming of multiple bicycle delivery business district sets according to the constraint conditions of high-density delivery specifically includes:

[0011] Select the logistics center as the reference point or select the business district farthest from the logistics center as the reference point;

[0012] Taking the business district farthest from the reference point as the starting point, cluster the business districts around the starting point according to the constraints of high-density delivery, and divide the vehicles according to the clustering results to form the first set of single-bike delivery business districts;

[0013] In the remaining business districts, continue to use the business district farthest from the reference point as the starting point, cluster the business districts around the starting point according to the constraints of high-density distribution, and divide the vehicles according to the clustering results to form the second set of single-bike delivery business districts; repeat the above clustering process until all business districts are clustered to form multiple sets of single-bike delivery business districts.

[0014] On the basis of the above technical solution, preferably, the constraints of the high-density delivery include:

[0015] If any one of the three conditions, namely, the maximum load of a bicycle, the longest working time of a bicycle, and the maximum number of households that a bicycle can deliver to, is met, the conditions for vehicle division are met, and the clustering results are output for vehicle division.

[0016] On the basis of the above technical solutions, preferably, the delivery route planning of each transport vehicle according to the set of multiple bicycle delivery business circles specifically includes:

[0017] The total number of transport vehicles required and the delivery range of a single transport vehicle are determined based on multiple single-vehicle delivery business district sets; one transport vehicle is responsible for one single-vehicle delivery business district set;

[0018] Within the delivery range of a single transport vehicle, a dynamic route planning algorithm is used to plan the delivery path of a single transport vehicle.

[0019] Based on the above technical solution, preferably, the generating of the area route including all business districts according to the dynamic route optimization algorithm specifically includes:

[0020] All business districts are grouped into a cluster area, with the logistics center as the starting point and the end point. A dynamic route optimization algorithm is used in the cluster area to generate a shortest route that includes all business districts, and the shortest route is used as the cluster area route.

[0021] On the basis of the above technical solutions, preferably, the distribution route planning of each transport vehicle according to the collection area line specifically includes:

[0022] According to the real-time order volume of each business district, the collection area routes are intercepted and allocated to each transport vehicle in sequence according to the maximum load capacity of the transport vehicle and the longest working time of the transport vehicle, and the distribution path of each transport vehicle is generated;

[0023] On the basis of the above technical solution, preferably, the dynamic route optimization algorithm specifically includes:

[0024] The hyperparameters in the ant colony algorithm are learned through convolutional neural networks to obtain an improved ant colony algorithm.

[0025] Each path data planned by manual experience is used as the initial ant position, and the driving routes between different business districts are planned using the improved ant colony algorithm.

[0026] In a second aspect, the present invention discloses an intelligent planning system for tobacco logistics under high-density distribution constraints, the system comprising:

[0027] Retailer clustering module: used to cluster the geographical locations of retailers to be delivered to form multiple business districts;

[0028] Bicycle delivery collection module: It is used to form multiple bicycle delivery business district collections based on business districts and according to the constraints of high-density delivery;

[0029] Large route planning module: It is used to generate area routes that include all business districts based on the business district as the basic unit according to the dynamic route optimization algorithm;

[0030] Bicycle route planning module: Plan the delivery routes for each transport vehicle based on multiple bicycle delivery business district collections or area routes.

[0031] In a third aspect of the present invention, an electronic device is disclosed, comprising: at least one processor, at least one memory, a communication interface and a bus;

[0032] Wherein, the processor, memory, and communication interface communicate with each other via the bus;

[0033] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the method according to the first aspect of the present invention.

[0034] According to a fourth aspect of the present invention, a computer-readable storage medium is disclosed, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement the method described in the first aspect of the present invention.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1) The present invention proposes a slicing and car allocation method, which uses a clustering algorithm to organize the retailers to be delivered into multiple business districts, and then clusters the business districts according to the constraints of high-density delivery. In the process of business district clustering, load and time constraints are added. If any one of the three conditions of the maximum load of a single vehicle, the longest working time of a single vehicle, and the maximum number of households to be delivered by a single vehicle is met, the car allocation condition is met, and the slicing and car allocation is performed to form multiple single-vehicle delivery business district sets. One transport vehicle is responsible for one single-vehicle delivery business district set, which can indirectly improve the loading rate and realize efficient tobacco logistics loading;

[0037] 2) The present invention also proposes a route segmentation method based on the business district, which generates a collection area route including all business districts according to the dynamic route optimization algorithm. According to the real-time order quantity of each business district, the maximum loading capacity and the longest working time of the transport vehicle in the collection area route are intercepted and allocated to each transport vehicle in turn, and the distribution path of each transport vehicle is generated. This loading mechanism successfully balances the workload of each vehicle and improves the loading rate and distribution efficiency.

[0038] 3) The present invention can be applied to a large number of constraints in actual situations, so that the workload of each vehicle is even, the loading rate is high and the distribution efficiency is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of an intelligent planning method for tobacco logistics under high-density distribution constraints provided by an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of a collection of multiple bicycle delivery business districts formed with a logistics center as a reference point provided by an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of a collection of multiple bicycle delivery business districts formed by taking the business district farthest from the logistics center as a reference point provided in an embodiment of the present invention;

[0043] Figure 4 A schematic diagram of planning results of distribution route planning for each transport vehicle according to the collection area lines provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Embodiment 1

[0046] See also Figure 1 The present invention provides a tobacco logistics intelligent planning method under high-density distribution condition constraints, which performs logistics intelligent planning by slicing and dividing vehicles, and includes the following steps:

[0047] S11. Cluster and divide the geographical locations of the retailers to be delivered to form multiple business districts.

[0048] Based on the geographic information of the retailers to be delivered and the maximum working mileage of the transport vehicle, the order destinations with a time of less than or equal to 1 minute are clustered into a business district, which serves as the basic unit of the subsequent path planning algorithm.

[0049] S12. Taking the business district as the basic unit, form multiple bicycle delivery business district sets according to the constraints of high-density distribution.

[0050] The present invention divides the area that each transport vehicle is responsible for into segments based on the condition of the lowest delivery cost, so that one transport vehicle is responsible for one area, that is, a single-vehicle delivery business district set.

[0051] Step S12 specifically includes the following sub-steps:

[0052] S121. First, select a reference point (AP). There are two options: select the logistics center as the reference point or select the business district farthest from the logistics center as the reference point.

[0053] S122. Taking the business district farthest from the reference point as the starting point, clustering the business districts around the starting point according to the constraints of high-density delivery, and dividing the vehicles according to the clustering results to form the first set of single-bike delivery business districts.

[0054] Among them, clustering the business districts around the starting point is to cluster the business district locations, and the business districts closer to the starting point are divided into one category. After each business district cluster is completed, it is determined whether the constraints of high-density distribution are met. If the constraints of high-density distribution are met, the current clustering is stopped and a clustering result is obtained, which is a set of single-bike distribution business districts. The constraints of high-density distribution include: the maximum load of a single vehicle, the longest working time of a single vehicle, and the maximum number of households that a single vehicle can deliver. If any one of these three conditions is met, the vehicle division condition is met, the clustering is stopped, and the vehicle is divided according to the clustering result, that is, the single-bike distribution business district set obtained by the current clustering is responsible for delivery by one transport vehicle.

[0055] S123. In the remaining business districts, continue to use the business district farthest from the reference point as the starting point, cluster the business districts around the starting point according to the constraints of high-density distribution, and divide the vehicles according to the clustering results to form a second set of single-bike delivery business districts; repeat the above clustering process until all business districts are clustered to form multiple sets of single-bike delivery business districts.

[0056] like Figure 2 , 3 As shown, the white circle represents the business district. Figure 2 It is a collection of multiple bicycle delivery business districts formed with the logistics center as the reference point. Figure 3 A collection of multiple bicycle delivery business districts formed with the business district farthest from the logistics center as the reference point.

[0057] S13. Plan the delivery routes for each transport vehicle based on a set of multiple bicycle delivery business districts.

[0058] S13 specifically includes the following sub-steps:

[0059] S131. Determine the total number of transport vehicles required and the delivery range of a single transport vehicle based on a set of multiple single-vehicle delivery business districts.

[0060] like Figure 2 , Figure 3 As shown in , one transport vehicle is responsible for one bicycle delivery business district set. The total number of transport vehicles required for three bicycle delivery business district sets is 3. The delivery range of the i-th transport vehicle is determined by the i-th bicycle delivery business district set, i=1,2,3.

[0061] S132. Within the delivery range of a single transport vehicle, a dynamic route planning algorithm is used to plan the delivery route of the single transport vehicle.

[0062] The dynamic route optimization algorithm adopts an improved ant colony algorithm. Specifically, the hyperparameters in the ant colony algorithm, such as heuristic factors, pheromone parameters, etc., are learned through a convolutional neural network to better set the hyperparameters and facilitate rapid convergence to obtain an improved ant colony algorithm.

[0063] Dynamic route optimization within the delivery range of a single transport vehicle can be performed without returning to the starting point, that is, the initial positions of all simulated transport tool variables are set at a fixed starting point. Therefore, the present invention uses each artificial experience-planned path data as the initial ant position, and plans the driving routes between different business districts through the improved ant colony algorithm.

[0064] The principle of the ant colony algorithm is to first initialize the path variables. The number of variables increases each round. The corresponding state transition probability is calculated for each variable to select the next element, and a global taboo table is modified to continuously optimize the result. When a variable reaches the target of the loop, it will leave an information parameter in the element transfer process, which will instruct the next variable to perform a better state transfer process. Of course, the information parameter will also be gradually reduced according to the pre-set rules to prevent the emergence of local optimal solutions or fall into wrong solutions due to random initialization problems. Finally, with continuous iterations, the algorithm will provide a solution with the shortest distance, that is, the optimal solution.

[0065] The present invention clusters business districts according to the constraints of high-density distribution, adds load and time constraints in the process of business district clustering, and satisfies any one of the three conditions of maximum load capacity of a single vehicle, maximum working time of a single vehicle, and maximum number of households that a single vehicle can deliver to, thus meeting the vehicle division conditions. The vehicle is sliced ​​and divided to form multiple single-vehicle delivery business district sets, with one transport vehicle responsible for one single-vehicle delivery business district set, which can indirectly improve the loading rate and realize efficient tobacco logistics loading.

[0066] Embodiment 2

[0067] As an alternative embodiment to the first embodiment, the present invention provides another tobacco logistics intelligent planning method under high-density distribution condition constraints, which performs logistics intelligent planning by route segmentation, including the following steps:

[0068] S21. Cluster the geographical locations of the retailers to be delivered to form multiple business districts.

[0069] This process is the same as step S11.

[0070] S22. Generate a collection area route including all business districts according to a dynamic route optimization algorithm.

[0071] All business districts are grouped into a cluster area, and route planning is performed within the entire cluster area according to the dynamic route optimization algorithm.

[0072] Dynamic route optimization in the entire cluster area can be performed by returning to the starting point, that is, starting from the logistics center, traversing all business districts, and finally returning to the logistics center. The aforementioned improved ant colony algorithm is used to perform dynamic route optimization in the cluster area. The improved ant colony algorithm learns the hyperparameters in the ant colony algorithm through a convolutional neural network, and finally generates a shortest route that includes all business districts. The shortest route is used as the cluster area route.

[0073] S23. Plan the delivery routes for each transport vehicle according to the collection area routes.

[0074] According to the real-time order quantity of each business district, the area routes are intercepted and allocated to each transport vehicle in sequence according to the maximum load capacity of the transport vehicle and the longest working time of the transport vehicle, generating the distribution path for each transport vehicle.

[0075] Specifically, the above-mentioned high-density distribution constraints can be used to intercept and allocate them to each transport vehicle in turn to generate the distribution path for each transport vehicle. That is, the distribution path is planned in accordance with the collection area line in turn. When any one of the three conditions of the maximum load of a single vehicle, the maximum working time of a single vehicle, and the maximum number of households delivered by a single vehicle is met, the current collection area line is intercepted as the distribution path for the current transport vehicle.

[0076] like Figure 4 The figure shows the planning results of the distribution route planning of each transport vehicle according to the collection area line, where the delivery sequence number is the business district sequence number of the collection area line. The first transport vehicle (the first vehicle) is arranged for delivery according to the collection area line. When the first transport vehicle reaches the maximum load or the working time is greater than 8 hours or the maximum number of households that can be delivered by a single vehicle is reached, the second transport vehicle (the second vehicle) is immediately arranged to continue delivery according to the remaining collection area lines until all business districts are delivered.

[0077] Embodiment 3

[0078] Corresponding to the above method embodiment, the present invention proposes a tobacco logistics intelligent planning system under high-density distribution condition constraints, the system comprising:

[0079] Business district clustering module: used to cluster the geographical locations of retailers to be delivered to form multiple business districts;

[0080] Bike collection module: It is used to form multiple bike distribution business district collections based on business districts and according to the constraints of high-density distribution.

[0081] Large route planning module: It is used to generate area routes that include all business districts based on the business district as the basic unit according to the dynamic route optimization algorithm;

[0082] Bicycle route planning module: Plan the delivery routes for each transport vehicle based on multiple bicycle delivery business district collections or area routes.

[0083] The above system embodiments and method embodiments correspond to each other, and for a brief description of the system embodiments, please refer to the method embodiments.

[0084] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus; wherein the processor, memory, and communication interface communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the aforementioned method of the present invention.

[0085] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, wherein the computer instructions enable the computer to implement all or part of the steps of the method described in the embodiment of the present invention. The storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0086] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be distributed to multiple network units. A person skilled in the art may select some or all of the modules according to actual needs to achieve the purpose of the solution of this embodiment without creative effort.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A tobacco logistics intelligent planning method under high-density distribution constraints, characterized in that: The method comprises: Cluster the geographical locations of retailers to be delivered to form multiple business districts; Taking the business district as the basic unit, multiple single-bike delivery business district sets are formed according to the constraints of high-density delivery, or a collection area route including all business districts is generated according to the dynamic route optimization algorithm; Plan the delivery routes for each transport vehicle based on multiple single-vehicle delivery business district collections or area collection routes; The forming of multiple bicycle delivery business district sets according to the constraint conditions of high-density delivery specifically includes: Select the logistics center as the reference point or select the business district farthest from the logistics center as the reference point; Taking the business district farthest from the reference point as the starting point, cluster the business districts around the starting point according to the constraints of high-density delivery, and divide the vehicles according to the clustering results to form the first set of single-bike delivery business districts; In the remaining business districts, the business district farthest from the reference point is used as the starting point. Based on the constraints of high-density delivery, the business districts around the starting point are clustered. The vehicles are divided according to the clustering results to form the second set of single-bike delivery business districts. The above clustering process is repeated until all business districts are clustered to form multiple sets of single-bike delivery business districts. The constraints of high-density distribution include: If any one of the three conditions, namely, the maximum load of a bicycle, the longest working time of a bicycle, and the maximum number of households that a bicycle can deliver to, is met, the conditions for vehicle division are met, and the clustering results are output for vehicle division; The distribution route planning for each transport vehicle based on the collection of multiple bicycle distribution business circles specifically includes: The total number of transport vehicles required and the delivery range of a single transport vehicle are determined based on multiple single-vehicle delivery business district sets; one transport vehicle is responsible for one single-vehicle delivery business district set; Within the delivery range of a single transport vehicle, a dynamic route optimization algorithm is used to plan the delivery path of a single transport vehicle; The generating of a collection area route including all business districts according to the dynamic route optimization algorithm specifically includes: All business districts are grouped into a cluster area, with the logistics center as the starting point and the end point, and a dynamic route optimization algorithm is used in the cluster area to generate a shortest route that includes all business districts, and the shortest route is used as the cluster area route; The distribution route planning for each transport vehicle according to the collection area line specifically includes: According to the real-time order volume of each business district, the collection area routes are intercepted and allocated to each transport vehicle in sequence according to the maximum load capacity of the transport vehicle and the longest working time of the transport vehicle, and the distribution path of each transport vehicle is generated; The dynamic route optimization algorithm specifically includes: The hyperparameters in the ant colony algorithm are learned by a convolutional neural network to obtain an improved ant colony algorithm; the hyperparameters include a heuristic factor and a pheromone parameter; Each path data planned by manual experience is used as the initial ant position, and the driving routes between different business districts are planned using the improved ant colony algorithm.

2. A tobacco logistics intelligent planning system under high-density distribution condition constraints, which uses the tobacco logistics intelligent planning method under high-density distribution condition constraints as claimed in claim 1, characterized in that: The system comprises: Business district clustering module: used to cluster the geographical locations of retailers to be delivered to form multiple business districts; Bike collection module: It is used to form multiple bike distribution business district collections based on business districts and according to the constraints of high-density distribution. Large route planning module: It is used to generate area routes that include all business districts based on the business district as the basic unit according to the dynamic route optimization algorithm; Bicycle route planning module: Plan the delivery routes for each transport vehicle based on multiple bicycle delivery business district collections or area routes.

3. An electronic device, characterized in that: include: at least one processor, at least one memory, a communication interface, and a bus; Wherein, the processor, memory, and communication interface communicate with each other via the bus; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the method according to claim 1.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions cause a computer to implement the method of claim 1.

Images