Intelligent distribution scheduling method and intelligent distribution scheduling system

Abstract

The invention relates to the technical field of distribution scheduling, and discloses an intelligent distribution scheduling method and an intelligent distribution scheduling system, and the method comprises the steps: carrying out the regional grouping of stores according to the position information of the stores and historical grouping experience data, and flexibly carrying out the cross-group transfer of the stores according to the actual distribution order condition. And finally, through combination of an improved ant colony algorithm and a simulated annealing algorithm, multi-target optimization of a distribution path is realized, rich scheduling experience accumulated in an actual operation process is fully utilized, a scheduling decision is more suitable for an actual operation scene, a complex and changeable logistics task can be better coped with, the distribution efficiency is improved, and the distribution cost is reduced. According to the method, distribution of distribution tasks is optimized, waste of distribution resources is reduced, multiple targets of distribution mileage, cost, time and the like can be comprehensively considered, and scientificity and accuracy of distribution scheduling are effectively improved, so that the warehouse logistics industry is promoted to develop in a more efficient and more intelligent direction, and the efficient and accurate development requirements of modern logistics are met.

Description

Technical Field

[0001] This invention relates to the field of delivery scheduling technology, and in particular to an intelligent delivery scheduling method and an intelligent delivery scheduling system. Background Technology

[0002] In the current operational scenarios of the warehousing and logistics industry, the delivery and scheduling process still largely relies on human experience. This reliance on human experience for scheduling decisions not only proves inadequate when handling complex and ever-changing logistics tasks, but also results in low overall scheduling efficiency due to the limitations of manual operation, making it difficult to meet the demands of modern logistics for high efficiency and precision.

[0003] Although some intelligent delivery scheduling solutions have emerged in the industry, attempting to improve the scientific nature and efficiency of delivery scheduling through technological means, these solutions still have many shortcomings in practical applications. Specifically, existing intelligent delivery scheduling solutions often fail to fully value and effectively learn from and utilize the rich scheduling experience accumulated in actual operations. This valuable experience is a key element in improving scheduling quality, yet it is overlooked by most intelligent delivery scheduling solutions.

[0004] Furthermore, most existing intelligent delivery scheduling solutions focus on single-objective optimization, with a relatively singular optimization direction, such as pursuing only the shortest distance or the lowest cost. This single-objective optimization model makes it difficult for intelligent scheduling solutions to comprehensively and systematically solve the complex problems faced by warehousing and logistics delivery scheduling in practical applications.

[0005] In view of the various problems existing in the above-mentioned delivery scheduling solutions, in order to promote the development of the warehousing and logistics industry towards a more efficient and intelligent direction, it is urgent to carry out in-depth improvement and optimization of existing technologies in order to build a more scientific, comprehensive and intelligent delivery scheduling system.

[0006] The above information is provided as background information only to aid in understanding the present invention, and does not constitute an assertion or admission that any of the above content can be used as prior art relative to the present invention. Summary of the Invention

[0007] This invention provides an intelligent delivery scheduling method and an intelligent delivery scheduling system to solve the problems existing in the prior art.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] In a first aspect, the present invention provides an intelligent delivery scheduling method, the method comprising:

[0010] S101. Based on the location information and historical grouping experience data of each store, group each store by region to obtain several regional groups; each regional group includes several stores.

[0011] S102. Upon receiving a delivery order request, obtain the location information and volume of each store to be delivered in the order;

[0012] S103. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, determine whether it is necessary to transfer the stores to be delivered across groups; if yes, then execute S104; if no, then execute S105.

[0013] S104. Perform cross-group transfer of stores awaiting delivery to achieve regrouping;

[0014] S105. No cross-group transfer of stores awaiting delivery;

[0015] S106. Combining the improved ant colony algorithm and simulated annealing algorithm, multi-objective optimization of delivery paths is performed for each of the aforementioned regional groups to be delivered, so as to complete the intelligent scheduling of delivery.

[0016] Furthermore, in the intelligent delivery scheduling method, step S101 includes:

[0017] S1011. Obtain the location information of each store;

[0018] S1012. Clean and standardize the location information of each store;

[0019] S1013. Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results;

[0020] S1014. Based on historical grouping experience data, adjust the preliminary grouping results to obtain several regional groups; each regional group includes several stores.

[0021] Furthermore, in the intelligent delivery scheduling method, step S103 includes:

[0022] S1031. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups;

[0023] S1032. Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer conditions are that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state; if yes, then execute S1033; if no, then execute S1034.

[0024] S1033, Determine the cross-group transfer of stores that need to be delivered, and proceed to S104;

[0025] S1034. Determine which stores do not require cross-group transfer for delivery and proceed to S105.

[0026] Furthermore, in the intelligent delivery scheduling method, step S104 includes:

[0027] S1041. At least two of the regions that meet the transfer conditions are grouped together for cross-group transfer of stores to be delivered. The goal of the cross-group transfer is that the total volume of goods to be delivered in each of the region groups can be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state.

[0028] Furthermore, in the intelligent delivery scheduling method, step S106 includes:

[0029] S1061. Based on the improved ant colony algorithm, perform multi-objective optimization of delivery paths for each of the areas to be delivered, and obtain several corresponding preliminary planned paths.

[0030] S1062. Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to complete the intelligent scheduling of delivery.

[0031] Secondly, the present invention provides an intelligent delivery scheduling system, the system comprising:

[0032] The regional grouping module is used to group each store by region based on its location information and historical grouping experience data, resulting in several regional groups; each regional group includes several stores.

[0033] The information acquisition module is used to acquire the location information and volume of each store to be delivered in the order when a delivery order request is received;

[0034] The transfer judgment module is used to determine whether a cross-group transfer of stores to be delivered is necessary based on the location information of each store to be delivered and the corresponding volume of goods to be delivered.

[0035] The transfer and reorganization module is used to perform cross-group transfers of stores awaiting delivery if cross-group transfers are required, thereby achieving regrouping; otherwise, cross-group transfers of stores awaiting delivery are not performed.

[0036] The path optimization module combines an improved ant colony algorithm and a simulated annealing algorithm to perform multi-objective optimization of delivery paths for each of the area groups to be delivered, thereby completing intelligent delivery scheduling.

[0037] Furthermore, in the intelligent delivery scheduling system, the regional grouping module is specifically used for:

[0038] Obtain the location information of each store;

[0039] Clean and standardize the location information of each store;

[0040] Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results;

[0041] Based on historical grouping experience data, the initial grouping results are adjusted to obtain several regional groups; each regional group includes several stores.

[0042] Furthermore, in the intelligent delivery scheduling system, the transfer judgment module is specifically used for:

[0043] Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups;

[0044] Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer conditions are that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state.

[0045] If yes, then it is determined that the cross-group transfer of the stores awaiting delivery needs to be carried out; if no, then it is determined that the cross-group transfer of the stores awaiting delivery does not need to be carried out.

[0046] Furthermore, in the intelligent delivery scheduling system, the transfer and reorganization module is specifically used for:

[0047] The stores to be delivered are transferred across groups of at least two of the regions that meet the transfer conditions. The goal of the cross-group transfer is to ensure that the total volume of goods to be delivered in each of the regional groups can be carried by a certain number of vehicles at full or near full capacity.

[0048] Furthermore, in the intelligent delivery scheduling system, the route optimization module is specifically used for:

[0049] Based on the improved ant colony algorithm, multi-objective optimization of delivery paths is performed for each of the areas to be delivered, resulting in several preliminary planned paths.

[0050] Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to achieve intelligent scheduling of delivery.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] This invention provides an intelligent delivery scheduling method and system. By grouping stores by region based on their location information and historical grouping experience data, and then flexibly transferring stores across groups according to actual delivery orders, the system combines improved ant colony optimization and simulated annealing algorithms to achieve multi-objective optimization of delivery routes. This not only fully utilizes the rich scheduling experience accumulated in actual operations, making scheduling decisions more aligned with real-world operational scenarios and better able to handle complex and ever-changing logistics tasks, but also improves delivery efficiency, optimizes the allocation of delivery tasks, and reduces waste of delivery resources. It comprehensively considers multiple objectives such as delivery mileage, cost, and time, effectively improving the scientific nature and accuracy of delivery scheduling. This, in turn, helps promote the development of the warehousing and logistics industry towards greater efficiency and intelligence, meeting the needs of modern logistics for high efficiency and precision.

[0053] The present invention has other features and advantages, which will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart illustrating an intelligent delivery scheduling method provided in Embodiment 1 of the present invention;

[0056] Figure 2 This is a further detailed flowchart of S101 provided in Embodiment 1 of the present invention;

[0057] Figure 3 This is a further detailed flowchart of S103 provided in Embodiment 1 of the present invention;

[0058] Figure 4 This is a further detailed flowchart of S104 provided in Embodiment 1 of the present invention;

[0059] Figure 5 This is a further detailed flowchart of S106 provided in Embodiment 1 of the present invention;

[0060] Figure 6This is a schematic diagram of the functional modules of an intelligent delivery scheduling system provided in Embodiment 2 of the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0062] Example 1

[0063] Please refer to Figure 1 This is a flowchart illustrating an intelligent delivery scheduling method provided in Embodiment 1 of the present invention. This method is applicable to scenarios involving delivery scheduling for stores with pending deliveries. The method is executed by an intelligent delivery scheduling system, which can be implemented using software and / or hardware. The method specifically includes the following steps:

[0064] S101. Based on the location information and historical grouping experience data of each store, group each store by region to obtain several regional groups; each regional group includes several stores.

[0065] It should be noted that this step, by grouping stores by region, provides a basic framework for subsequent delivery scheduling, enabling more organized regional management of delivery tasks, facilitating centralized resource allocation, and improving delivery efficiency.

[0066] The specific operation involves the intelligent delivery scheduling system collecting location information for each store, such as latitude and longitude coordinates and address information, while also referencing historical grouping experience data. This historical data may include past performance in delivery efficiency, cost, and other aspects under different regional groups. Based on this information, the system uses specific grouping algorithms or rules to group stores that are geographically close or have similar delivery characteristics into the same regional group, ultimately forming several such regional groups, each containing a certain number of stores.

[0067] Suppose a city has stores in multiple areas. The system can group stores located in the city center business district into one group and stores located in the suburban industrial area into another group, based on information such as the street and business district where the stores are located, and combined with which stores have better delivery performance in the past (such as shorter delivery time and lower cost).

[0068] S102. Upon receiving a delivery order request, obtain the location information and volume of each store to be delivered in the order.

[0069] It should be noted that this step involves real-time monitoring of key information about delivery orders, providing necessary data support for subsequent decisions on whether cross-group transfers are necessary and for optimizing delivery routes.

[0070] The specific operation can be as follows: When a new delivery order request enters the intelligent delivery dispatch system, the system extracts the specific location information of each store to be delivered from the order information, which can be data such as latitude and longitude, address information, etc., and at the same time obtains the volume information of the goods to be delivered for each store. This information reflects the actual demand and scale of this delivery task.

[0071] For example, if a delivery order requires delivery to three stores, the system will know that the three stores are located in the east, south and west of the city, and will also know the volume of the goods to be delivered to each store, such as 5 cubic meters for the first store, 3 cubic meters for the second, and 4 cubic meters for the third.

[0072] S103. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, determine whether it is necessary to transfer the stores to be delivered across groups; if yes, then execute S104; if no, then execute S105.

[0073] It should be noted that this step flexibly adjusts the grouping of the store's area based on real-time order status to optimize the delivery plan, ensure that delivery resources are allocated more rationally, and improve overall delivery efficiency.

[0074] The specific operation can be as follows: The system analyzes the location information and volume of goods to be delivered based on the obtained information from the stores to be delivered, using preset judgment rules or algorithms. For example, it considers whether transferring a store from the current group to another group would shorten the overall delivery route, reduce costs, or make the delivery time more reasonable. If the judgment result indicates that a cross-group transfer is required, the system proceeds to step S104; if the judgment indicates that a cross-group transfer is not required, the system proceeds to step S105.

[0075] Suppose that system analysis finds that transferring a store originally belonging to group A to group B, combined with the store's cargo volume, can make the delivery vehicles in group B load more efficiently and reduce the overall delivery mileage, then it is determined that a cross-group transfer is necessary; conversely, if analysis finds that a cross-group transfer will not bring significant delivery optimization effects, then it is determined that a cross-group transfer is not necessary.

[0076] S104. Perform cross-group transfer of stores awaiting delivery to achieve regrouping.

[0077] It should be noted that when it is determined that cross-group transfer is necessary, the allocation of delivery resources can be optimized by adjusting the grouping of stores, so that the delivery plan is more in line with the actual order demand and delivery efficiency is improved.

[0078] The specific operation can be as follows: Based on the judgment result of S103, the system removes the stores that need to be transferred across groups from their original groups and adds them to the target group, thus completing the regrouping operation of the stores. This process may involve updating group information in the system, the relationship between stores and groups, and other data.

[0079] For example, if the system determines that a store in group A needs to be moved to group B, as in the example above, it will delete the store from the records in group A and add the store information to the records in group B, thus completing the regrouping.

[0080] S105. No cross-group transfer of stores awaiting delivery.

[0081] It should be noted that when it is determined that cross-group transfer is not necessary, the original store group status should be maintained, and the subsequent delivery scheduling process should continue according to the original group to avoid the operational complexity and potential risks caused by unnecessary adjustments.

[0082] The specific operation can be as follows: the system does not perform any operations related to the transfer of stores across groups, maintains the current composition of stores in each regional group, and directly proceeds to the subsequent delivery route optimization steps.

[0083] For example, if S103 determines that cross-group transfer is not required, the system will not make any store grouping adjustments and will directly proceed to step S106 to optimize the delivery route.

[0084] S106. Combining the improved ant colony algorithm and simulated annealing algorithm, multi-objective optimization of delivery paths is performed for each of the aforementioned regional groups to be delivered, so as to complete the intelligent scheduling of delivery.

[0085] It should be noted that this step uses advanced algorithms to optimize delivery routes, taking into account multiple objectives (such as delivery mileage, cost, time, etc.) to achieve intelligent delivery scheduling, improve delivery efficiency and scientific rigor, and reduce resource waste.

[0086] The specific operation involves: an improved ant colony algorithm simulating the behavior of ants searching for the optimal path during foraging, gradually searching for a better delivery path through the pheromone transmission and update mechanism of ants along the path. The simulated annealing algorithm borrows the principle of metal annealing, allowing the algorithm to accept some poor solutions at a certain temperature to avoid getting trapped in local optima, and gradually converging to the global optimum as the temperature decreases. The system combines these two algorithms, grouping each region and performing multi-objective optimization calculations for delivery paths, focusing on multiple objectives such as delivery mileage, cost, and time. Ultimately, it obtains the optimal delivery path for each region group, completing the entire intelligent scheduling process for delivery.

[0087] For example, for each region group after grouping (which may or may not involve cross-group transfers), the system uses an optimization model that combines an improved ant colony algorithm and a simulated annealing algorithm. It inputs data such as the location information and volume of goods in each group, and after multiple iterations, it obtains the optimal route for vehicles to travel in each group. This route comprehensively considers multiple factors such as how to minimize the total mileage of the vehicles, the total cost, and the delivery time.

[0088] Please refer to Figure 2 In one embodiment of this example, step S101 can be further refined to include the following steps: through a series of orderly operations, ensuring that store grouping is more scientific and reasonable, and fully considering the accuracy of data processing and the integration of actual operational experience:

[0089] S1011. Obtain the location information of each store.

[0090] It should be noted that location information is one of the core bases for grouping stores. Accurately obtaining the location data of each store can provide basic and crucial information support for subsequent grouping operations, making the grouping results more consistent with the actual geographical distribution.

[0091] Specifically, the intelligent delivery dispatch system can collect the location information of each store through various means. For example, it can interact with the store's electronic devices or management system to directly obtain its recorded geographical coordinates (such as latitude and longitude) or address information; or it can use Geographic Information System (GIS) technology to combine the store's address information and convert it into precise coordinate data.

[0092] Suppose a chain supermarket has multiple stores. The system connects to the POS system or store management backend of each store to obtain the latitude and longitude coordinates of each store on a map, such as store A's coordinates being (116.404, 39.915) and store B's coordinates being (116.397, 39.908), etc. Alternatively, it can obtain the address information of each store, such as store A's address being Daohuacun Road, Nancheng Street, Dongguan City, and store B's address being Qifeng Road, Dongcheng Street, Dongguan City.

[0093] S1012. Clean and standardize the location information of each store.

[0094] It should be noted that the raw location information may contain various problems, such as inconsistent data formats, errors, or noisy data. Cleaning and standardization processes can improve data quality and consistency, ensuring that subsequent grouping algorithms are based on accurate and standardized data, and avoiding biases in grouping results due to data issues.

[0095] The specific steps are as follows:

[0096] Data cleaning: Check for outliers in location information, such as the coordinates of some stores being significantly outside the reasonable range (e.g., negative coordinates and not within the normal geographical area); handle missing values, if the location information of a store is missing, it can be supplemented by the average location of surrounding stores or by address-based relocation; remove duplicate data to avoid the location information of the same store being recorded multiple times.

[0097] Data standardization: Standardize the format of location information, such as using a decimal latitude and longitude representation and specifying the number of decimal places, so that the location data of all stores are consistent in format, which facilitates subsequent algorithm processing.

[0098] For example, in the location information obtained, the coordinates of a certain store were found to be (116x, 39y), which was obviously a format error. After data cleaning, it was corrected to the correct latitude and longitude format (116.xxx, 39.xxx). At the same time, the coordinate data of all stores were uniformly retained to six decimal places to achieve data standardization.

[0099] S1013. Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results.

[0100] It should be noted that this step utilizes community detection algorithms to uncover the characteristics of potential community structures (i.e., regional grouping) in the data. Based on the location information of the stores, stores that are geographically close or closely related are initially grouped together, providing a basic framework for subsequent adjustments to the grouping based on historical experience.

[0101] The specific operation can be as follows: There are various community detection algorithms, such as the Louvain algorithm and the GN algorithm. Taking the Louvain algorithm as an example, this algorithm continuously optimizes the community structure by calculating the modularity (an indicator that measures the quality of community division). The system takes the cleaned and standardized store location information as input, and the algorithm divides the stores into several communities (i.e., preliminary grouping) based on factors such as the distance between stores and the tightness of their connections, so that the stores in each community are relatively concentrated geographically.

[0102] For example, suppose there are 20 stores. After processing by the community detection algorithm, these 20 stores are initially divided into 4 communities. Each community contains a different number of stores, such as community 1 with 6 stores, community 2 with 5 stores, community 3 with 4 stores, and community 4 with 5 stores.

[0103] S1014. Based on historical grouping experience data, adjust the preliminary grouping results to obtain several regional groups; each regional group includes several stores.

[0104] It should be noted that because the community detection algorithm only analyzes current location information, it may overlook the rich experience accumulated during actual operation. Adjusting the initial grouping results by combining historical grouping experience data can make the grouping results more in line with actual delivery needs and make full use of the successful experiences in delivery efficiency, cost, etc. in previous grouping.

[0105] The specific operation involves: The system collecting historical grouping data, including metrics such as delivery mileage, cost, and time under different groups in the past, as well as problems encountered and optimization solutions during the delivery process. This historical data is then compared and analyzed with the initial grouping results. For example, if historical data shows that placing certain stores in the same group results in higher delivery efficiency, but these stores were assigned to different groups in the initial grouping, then these stores are moved to the same group; conversely, if certain stores frequently cause excessively high delivery costs in historical groupings, but are grouped together in the initial grouping, then they are considered for separation. Through this adjustment, several final regional groups are obtained.

[0106] Suppose that in the initial grouping, store C and store D were assigned to different communities. However, historical grouping data shows that when these two stores were in the same group for deliveries, the vehicle load rate was higher and the delivery distance was shorter. Therefore, after adjusting based on historical grouping experience data, store C and store D were moved to the same regional group, resulting in the final grouping result. Each group contains several stores to meet actual delivery needs.

[0107] Please refer to Figure 3 In one embodiment of this example, step S103 can be further refined to include the following steps, which aim to provide clear guidance for subsequent delivery scheduling by accurately calculating the regional grouping of goods and strictly judging the transfer conditions, thereby ensuring that the delivery plan is both efficient and economical:

[0108] S1031. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups.

[0109] It is important to clarify the total amount of goods to be delivered in each regional group, as this is the basis for determining whether vehicles can be fully or nearly fully loaded. Only by accurately grasping the cargo volume of each group can vehicle allocation be rationally planned, avoiding situations where vehicles are underloaded and waste transport capacity, or overloaded and unable to transport goods.

[0110] The specific operation can be as follows: The intelligent delivery scheduling system first obtains the location information and corresponding volume data of each store to be delivered from the delivery orders. Then, according to the previously divided regional groups, the volume of the goods to be delivered for all stores belonging to the same regional group is added up. For example, if regional group 1 includes stores A, B, and C, and the volume of the goods to be delivered for store A is 5 cubic meters, store B is 3 cubic meters, and store C is 2 cubic meters, then the total volume of the goods to be delivered for regional group 1 is 5 + 3 + 2 = 10 cubic meters.

[0111] Suppose a chain supermarket is conducting distribution scheduling, with three regional groups. Regional group one contains 4 stores with goods volumes of 8 cubic meters, 6 cubic meters, 4 cubic meters, and 2 cubic meters respectively, totaling 8 + 6 + 4 + 2 = 20 cubic meters; regional group two contains 3 stores with goods volumes of 7 cubic meters, 5 cubic meters, and 3 cubic meters respectively, totaling 7 + 5 + 3 = 15 cubic meters; regional group three contains 2 stores with goods volumes of 9 cubic meters and 1 cubic meter respectively, totaling 9 + 1 = 10 cubic meters.

[0112] S1032. Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer condition is that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state; if yes, then execute S1033; if no, then execute S1034.

[0113] It should be noted that the purpose of this step is to determine whether there are regional groups that need to optimize vehicle allocation, that is, to determine whether there are multiple groups that need to transfer stores across groups because the cargo volume cannot be reasonably loaded into vehicles, so as to achieve more efficient use of vehicle resources.

[0114] The specific operation can be as follows: The system pre-sets the cargo capacity standard for vehicles, for example, the maximum cargo capacity of each vehicle is 12 cubic meters, and near full load can be set to a cargo capacity of 10 cubic meters or more. Then, the total volume of goods to be delivered in each area group is compared with the vehicle's cargo capacity standard. If the total volume of goods in an area group cannot be transported by a single vehicle at full capacity (less than 12 cubic meters but greater than the remaining amount that a single vehicle can reasonably carry, resulting in a low loading rate when dispatched alone), and cannot be combined with other small amounts of goods to reach full load or near full load (for example, 3 cubic meters of goods remaining, which is difficult to reasonably combine with other small amounts of goods to reach 10 cubic meters or more), and there are at least two such area groups, then the transfer conditions are met.

[0115] Continuing with the three regional groups mentioned above, assume the maximum cargo capacity of a vehicle is 12 cubic meters, and a near-full load is 10 cubic meters or more. The total cargo volume of Regional Group 1 is 20 cubic meters, which can be transported by two vehicles. The first vehicle carries 12 cubic meters, nearly full, while the second vehicle carries 8 cubic meters, a moderate loading rate. The total cargo volume of Regional Group 2 is 15 cubic meters. One vehicle carries 12 cubic meters, nearly full, and the remaining 3 cubic meters would have a low loading rate if a separate vehicle were used. The total cargo volume of Regional Group 3 is 10 cubic meters, which can be transported by one vehicle, nearly full. In this case, Regional Group 2 has a situation where the cargo volume cannot be reasonably loaded onto a vehicle, but since only one group is in this situation, it does not meet the condition that at least two groups meet the transfer criteria, so S1034 is executed. If there is another similar regional group with a total cargo volume of 11 cubic meters, one vehicle cannot fully load 12 cubic meters, and sending a separate vehicle is not cost-effective. In this case, two groups meet the transfer criteria, and S1033 is executed.

[0116] S1033, Determine the cross-group transfer of stores that need to be delivered, and proceed to S104.

[0117] It should be noted that when there are at least two regional groups that meet the transfer conditions, it indicates that the vehicle resources are not being fully utilized under the current grouping method. By transferring stores across groups, the distribution of goods can be readjusted, allowing vehicles to be loaded more rationally, improving delivery efficiency and reducing delivery costs.

[0118] The specific operation can be as follows: The system determines the stores that need to be transferred across groups based on certain rules (such as the distance between stores, the degree of matching of goods volume, etc.). For example, stores with smaller goods volume and closer distance to stores in other groups are transferred to groups with larger goods volume and lower vehicle loading rates to achieve a balanced distribution of goods volume. After determining the stores to be transferred, the system proceeds to step S104 to carry out subsequent cross-group transfer operations and formulate new delivery plans.

[0119] Assuming there are two groups meeting the transfer criteria, system analysis reveals that one store in Group 2 has a cargo volume of 3 cubic meters, and this store is relatively close to some stores in Group 1. Therefore, it is decided to transfer this 3-cubic-meter store to Group 1, increasing the cargo volume in Group 1 to 20 + 3 = 23 cubic meters, allowing for more efficient vehicle loading. The cargo volume in Group 2 becomes 15 - 3 = 12 cubic meters, which is just enough for one vehicle to be nearly fully loaded. Then, step S104 is executed to perform the cross-group transfer and update the delivery plan.

[0120] S1034. Determine which stores do not require cross-group transfer for delivery and proceed to S105.

[0121] It should be noted that when there are no at least two regional groups that meet the transfer conditions, it means that the vehicle resources can be used reasonably under the current grouping method. There is no need to transfer across groups, and the process can directly proceed to the subsequent delivery plan formulation and execution stage to improve the efficiency of the overall delivery process.

[0122] The specific operation can be as follows: After the system confirms that there is no need for cross-group transfer, it directly enters step S105, and formulates a specific delivery plan based on the existing regional grouping and cargo volume, including arranging vehicles and planning delivery routes.

[0123] As illustrated in the previous example, only area group two has a situation where the cargo volume cannot be reasonably loaded onto the vehicles, but since only one group is in this situation, it does not meet the transfer conditions. The system determines that no cross-group transfer is required, and then proceeds to step S105. Based on the existing three area group situations, suitable vehicles are assigned to each group for delivery, such as two vehicles for area group one, one vehicle for area group two, and one vehicle for area group three, and the corresponding delivery routes are planned.

[0124] Please refer to Figure 4 In one embodiment of this example, step S104 can be further refined to include the following steps to ensure that the total volume of goods to be delivered in each regional group can be reasonably fully or nearly fully loaded by the vehicle, thereby improving vehicle utilization and reducing delivery costs:

[0125] S1041. At least two of the regions that meet the transfer conditions are grouped together for cross-group transfer of stores to be delivered. The goal of the cross-group transfer is that the total volume of goods to be delivered in each of the region groups can be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state.

[0126] It should be noted that the core purpose of this step is to address the issue identified in S103 where unreasonable regional grouping of goods leads to vehicles being unable to operate at full capacity or near full capacity. By transferring goods between stores across groups and reallocating the goods, the goods volume of each regional group is matched with the vehicle's carrying capacity, thereby improving delivery efficiency, reducing empty runs and transportation trips, and lowering overall delivery costs.

[0127] The specific steps are as follows:

[0128] Determine the scope of stores eligible for transfer: The system first analyzes the information of all stores in at least two regional groups that meet the transfer criteria, including store location, volume of goods to be delivered, etc. For example, there are two regional groups A and B. Group A contains stores a1 (3 cubic meters) and a2 (5 cubic meters), while group B contains stores b1 (4 cubic meters) and b2 (2 cubic meters).

[0129] Screening potential relocation stores: Based on certain screening rules, such as store inventory size and distance from stores in other groups, identify stores suitable for cross-group relocation. For example, prioritize stores with smaller inventory sizes and closer proximity to stores in other groups as potential relocation targets. Assuming store a1 has a small inventory size and is close to some stores in group B, then a1 becomes a potential relocation store.

[0130] Simulated Transfer Effect: After identifying potential transfer stores, the system simulates the effect of transferring the store from one group to another. It calculates the total volume of goods to be delivered in both groups after the transfer, and determines whether a vehicle can be fully or nearly fully loaded based on vehicle load standards (e.g., a maximum load of 12 cubic meters per vehicle, with 10 cubic meters or more for near-full load). For example, after transferring store a1 (3 cubic meters) from group A to group B, the total volume of goods in group A becomes 5 cubic meters, and the total volume in group B becomes 4 + 2 + 3 = 9 cubic meters. At this point, group A can be accommodated by one vehicle, while group B is near full load but still not ideal, requiring further adjustments.

[0131] Optimize the transfer plan: If the simulated transfer effect is not ideal, the system continues to try transfer combinations of other stores or adjusts the number of stores to be transferred. For example, after transferring store a2 (5 cubic meters) from group A to group B, the cargo volume of group A becomes 0 (there are no stores waiting for delivery, and the actual plan may be re-planned), and the total cargo volume of group B becomes 4+2+3+5=14 cubic meters. One vehicle can be arranged to carry 12 cubic meters, which is close to full capacity. The remaining 2 cubic meters can be combined with other small cargo volumes or arranged for small vehicle transportation, resulting in a better overall effect. Through continuous trial and optimization, the best cross-group transfer plan is found.

[0132] Execute cross-group transfer: After determining the optimal transfer plan, the system officially executes the cross-group transfer operation, updating the store information and inventory data for each regional group. Simultaneously, the transfer process and related data are recorded for subsequent analysis and traceability.

[0133] Suppose there are three regional groups: C, D, and E. The total volume of goods to be delivered in group C is 8 cubic meters, in group D it is 15 cubic meters, and in group E it is 9 cubic meters. The maximum load capacity of a vehicle is 12 cubic meters, and it is close to full load at 10 cubic meters or more.

[0134] System analysis revealed that groups C and D met the transfer criteria. Group C contains stores c1 (3 cubic meters of goods) and c2 (5 cubic meters of goods); group D contains stores d1 (7 cubic meters of goods) and d2 (8 cubic meters of goods); and group E contains stores e1 (4 cubic meters of goods) and e2 (5 cubic meters of goods).

[0135] After screening and simulation, it was determined that store c2 (5 cubic meters) would be transferred from group C to group E. After the transfer, the total volume of goods in group C will be 3 cubic meters, which can be transported by one small vehicle; the total volume of goods in group E will become 4+5+5=14 cubic meters, which can be transported by one vehicle carrying 12 cubic meters, almost at full capacity, with the remaining 2 cubic meters to be handled flexibly; group D will remain at 15 cubic meters, which can be transported by one vehicle carrying 12 cubic meters, almost at full capacity, with the remaining 3 cubic meters to be combined with other goods for transportation.

[0136] The system executes the cross-group transfer operation, updates the information of each group, and makes the total volume of goods to be delivered in each regional group closer to the state of full or near full vehicle load, thereby improving vehicle utilization and delivery efficiency.

[0137] Please refer to Figure 5 In one embodiment of this invention, step S106 can be broken down into two closely linked steps: first, multi-objective optimization is performed using an improved ant colony algorithm to obtain a preliminary planned path; then, simulated annealing algorithm is used to make local fine-tuning adjustments to the preliminary path, ultimately achieving intelligent scheduling of delivery and ensuring that the delivery plan is both efficient and economical.

[0138] S1061. Based on the improved ant colony algorithm, perform multi-objective optimization of delivery paths for each of the regions to be delivered, and obtain several corresponding preliminary planned paths.

[0139] It should be noted that the ant colony algorithm has characteristics such as positive feedback and distributed computing, which can effectively simulate the process of finding the optimal path in actual delivery. The improved ant colony algorithm is further optimized on this basis, aiming to comprehensively consider multiple delivery objectives (such as shortest path, least time, lowest cost, etc.) and find multiple potential initial delivery paths for each region group, providing a rich foundation for subsequent fine-tuning.

[0140] S1062. Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to complete the intelligent scheduling of delivery.

[0141] It should be noted that the initial planned path obtained by the improved ant colony algorithm may contain local optima rather than global optima. The simulated annealing algorithm has the ability to escape local optima. By making local adjustments to the initial planned path, it further optimizes the path, making the final delivery plan more reasonable, improving delivery efficiency, and reducing delivery costs.

[0142] Although this invention frequently uses terms such as "regional grouping" and "cross-group transfer," the possibility of using other terms is not excluded. These terms are used merely for the convenience of describing and explaining the essence of this invention; interpreting them as any additional limitation would contradict the spirit of this invention.

[0143] This invention provides an intelligent delivery scheduling method. It groups stores by region based on their location information and historical grouping experience data, then flexibly transfers stores across groups according to actual delivery orders. Finally, by combining an improved ant colony algorithm and simulated annealing algorithm, it achieves multi-objective optimization of delivery routes. This method not only fully utilizes the rich scheduling experience accumulated in actual operations, making scheduling decisions more aligned with real-world operational scenarios and better able to handle complex and ever-changing logistics tasks, but also improves delivery efficiency, optimizes the allocation of delivery tasks, and reduces waste of delivery resources. It comprehensively considers multiple objectives such as delivery mileage, cost, and time, effectively improving the scientific nature and accuracy of delivery scheduling. This, in turn, helps promote the development of the warehousing and logistics industry towards greater efficiency and intelligence, meeting the needs of modern logistics for high efficiency and precision.

[0144] Example 2

[0145] Figure 6 This is a functional module diagram of an intelligent delivery scheduling system provided in Embodiment 2 of the present invention. This system is applicable to executing the intelligent delivery scheduling method provided in this embodiment of the present invention. The system specifically includes the following modules:

[0146] The regional grouping module 201 is used to group each store by region based on the location information and historical grouping experience data of each store, resulting in several regional groups; each regional group includes several stores;

[0147] The information acquisition module 202 is used to acquire the location information and volume of each store to be delivered in the order when a delivery order request is received;

[0148] The transfer judgment module 203 is used to determine whether it is necessary to transfer the stores to be delivered across groups based on the location information of each store to be delivered and the corresponding volume of goods to be delivered.

[0149] The transfer and reorganization module 204 is used to perform cross-group transfer of stores to be delivered if cross-group transfer of stores to be delivered is required, so as to achieve regrouping; if cross-group transfer of stores to be delivered is not required, cross-group transfer of stores to be delivered is not performed.

[0150] The path optimization module 205 is used to combine the improved ant colony algorithm and the simulated annealing algorithm to perform multi-objective optimization of delivery paths for each of the area groups to be delivered, so as to complete the intelligent scheduling of delivery.

[0151] Optionally, the region grouping module 201 is specifically used for:

[0152] Obtain the location information of each store;

[0153] Clean and standardize the location information of each store;

[0154] Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results;

[0155] Based on historical grouping experience data, the initial grouping results are adjusted to obtain several regional groups; each regional group includes several stores.

[0156] Optionally, the transfer determination module 203 is specifically used for:

[0157] Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups;

[0158] Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer conditions are that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state.

[0159] If yes, then it is determined that the cross-group transfer of the stores awaiting delivery needs to be carried out; if no, then it is determined that the cross-group transfer of the stores awaiting delivery does not need to be carried out.

[0160] Optionally, the transfer and recombination module 204 is specifically used for:

[0161] The stores to be delivered are transferred across groups of at least two of the regions that meet the transfer conditions. The goal of the cross-group transfer is to ensure that the total volume of goods to be delivered in each of the regional groups can be carried by a certain number of vehicles at full or near full capacity.

[0162] Optionally, the path optimization module 205 is specifically used for:

[0163] Based on the improved ant colony algorithm, multi-objective optimization of delivery paths is performed for each of the areas to be delivered, resulting in several preliminary planned paths.

[0164] Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to achieve intelligent scheduling of delivery.

[0165] The above system can execute the methods provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the methods.

[0166] Finally, it should be noted that although the above embodiments have been described in the description and drawings of this invention, this should not limit the scope of patent protection of this invention. Any technical solutions that are based on the essential concept of this invention, utilize the content described in the description and drawings of this invention to make equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this invention.

Claims

1. An intelligent delivery scheduling method, characterized in that, The method includes: S101. Based on the location information and historical grouping experience data of each store, group each store by region to obtain several regional groups; each regional group includes several stores. S102. Upon receiving a delivery order request, obtain the location information and volume of each store to be delivered in the order; S103. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, determine whether it is necessary to transfer the stores to be delivered across groups; if yes, then execute S104; if no, then execute S105. S104. Perform cross-group transfer of stores awaiting delivery to achieve regrouping; S105. No cross-group transfer of stores awaiting delivery; S106. Combining the improved ant colony algorithm and simulated annealing algorithm, multi-objective optimization of delivery paths is performed for each of the aforementioned regional groups to be delivered, so as to complete the intelligent scheduling of delivery.

2. The intelligent delivery scheduling method according to claim 1, characterized in that, S101 includes: S1011. Obtain the location information of each store; S1012. Clean and standardize the location information of each store; S1013. Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results; S1014. Based on historical grouping experience data, adjust the preliminary grouping results to obtain several regional groups; each regional group includes several stores.

3. The intelligent delivery scheduling method according to claim 1, characterized in that, S103 includes: S1031. Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups; S1032. Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer conditions are that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state; if yes, then execute S1033; if no, then execute S1034. S1033, Determine the cross-group transfer of stores that need to be delivered, and proceed to S104; S1034. Determine which stores do not require cross-group transfer for delivery and proceed to S105.

4. The intelligent delivery scheduling method according to claim 3, characterized in that, S104 includes: S1041. At least two of the regions that meet the transfer conditions are grouped together for cross-group transfer of stores to be delivered. The goal of the cross-group transfer is that the total volume of goods to be delivered in each of the region groups can be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state.

5. The intelligent delivery scheduling method according to claim 1, characterized in that, S106 includes: S1061. Based on the improved ant colony algorithm, perform multi-objective optimization of delivery paths for each of the areas to be delivered, and obtain several corresponding preliminary planned paths. S1062. Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to complete the intelligent scheduling of delivery.

6. An intelligent delivery scheduling system, characterized in that, The system includes: The regional grouping module is used to group each store by region based on its location information and historical grouping experience data, resulting in several regional groups; each regional group includes several stores. The information acquisition module is used to acquire the location information and volume of each store to be delivered in the order when a delivery order request is received; The transfer judgment module is used to determine whether a cross-group transfer of stores to be delivered is necessary based on the location information of each store to be delivered and the corresponding volume of goods to be delivered. The transfer and reorganization module is used to perform cross-group transfers of stores awaiting delivery if cross-group transfers are required, thereby achieving regrouping; otherwise, cross-group transfers of stores awaiting delivery are not performed. The path optimization module combines an improved ant colony algorithm and a simulated annealing algorithm to perform multi-objective optimization of delivery paths for each of the area groups to be delivered, thereby completing intelligent delivery scheduling.

7. The intelligent delivery scheduling system according to claim 6, characterized in that, The region grouping module is specifically used for: Obtain the location information of each store; Clean and standardize the location information of each store; Based on the location information of each store, a community detection algorithm is used to initially group each store by region to obtain preliminary grouping results; Based on historical grouping experience data, the initial grouping results are adjusted to obtain several regional groups; each regional group includes several stores.

8. The intelligent delivery scheduling system according to claim 6, characterized in that, The transfer determination module is specifically used for: Based on the location information of each store to be delivered and the corresponding volume of goods to be delivered, calculate the total volume of goods to be delivered for each of the aforementioned regional groups; Determine whether there are at least two regional groups that meet the transfer conditions; wherein, the transfer conditions are that the total volume of goods to be delivered in the regional groups cannot be carried by a certain number of vehicles in a fully loaded or nearly fully loaded state. If yes, then it is determined that the cross-group transfer of the stores awaiting delivery needs to be carried out; if no, then it is determined that the cross-group transfer of the stores awaiting delivery does not need to be carried out.

9. The intelligent delivery scheduling system according to claim 8, characterized in that, The transfer and recombination module is specifically used for: The stores to be delivered are transferred across groups of at least two of the regions that meet the transfer conditions. The goal of the cross-group transfer is to ensure that the total volume of goods to be delivered in each of the regional groups can be carried by a certain number of vehicles at full or near full capacity.

10. The intelligent delivery scheduling system according to claim 6, characterized in that, The path optimization module is specifically used for: Based on the improved ant colony algorithm, multi-objective optimization of delivery paths is performed for each of the areas to be delivered, resulting in several preliminary planned paths. Based on the simulated annealing algorithm, each of the initially planned paths is locally refined to achieve intelligent scheduling of delivery.

Images