A multi-objective vehicle route optimization method that takes into account both loading rate and transportation distance

Abstract

This invention discloses a multi-objective vehicle routing optimization method that considers both loading rate and transportation mileage. The method involves: acquiring order data of orders to be delivered and vehicle rated load and current location data; constructing an order load feature vector based on cargo weight and volume; clustering orders according to load differences to form multiple order clusters; assigning an initial priority label to each order cluster; sorting the order clusters according to the initial priority labels; merging the order clusters based on their spatial location information and loading rate constraints to generate an initial vehicle route; constructing a multi-objective collaborative optimization route evaluation model to optimize the initial vehicle route, generating a set of alternative route schemes; and selecting the route scheme with the highest comprehensive score as the final route scheme output. This invention can effectively reduce scheduling complexity, improve vehicle loading and delivery efficiency, and enhance the stability and practical feasibility of route decisions.

Description

Technical Field

[0001] This invention relates to the field of intelligent logistics scheduling technology, and in particular to a multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage. Background Technology

[0002] With the rapid development of e-commerce and urban logistics, the vehicle routing optimization problem has evolved from a single objective of minimizing transportation distance or cost to a complex optimization problem that comprehensively considers various practical constraints. Current research and applications widely employ modeling methods based on the Vehicle Routing Problem (VRP), combined with heuristic algorithms, intelligent optimization algorithms, or big data analytics to plan and adjust delivery routes. While these technologies have achieved some success in improving delivery efficiency and shortening transportation time, given the continuously increasing order volume and the growing diversity of cargo attributes, how to improve vehicle loading rates while ensuring transportation efficiency has become a crucial issue that urgently needs to be addressed in the field of logistics scheduling.

[0003] However, existing technologies generally suffer from insufficient characterization of order load features and simplistic path optimization objectives. On the one hand, some solutions focus on optimizing transportation mileage or time without fully considering the combined characteristics of cargo weight and volume, easily leading to wasted vehicle space or load-bearing resources and low overall loading rates. On the other hand, existing multi-objective path optimization methods often directly introduce complex objective functions during the path generation stage, lacking mechanisms for preprocessing and structured merging of order levels. This results in a large optimization search space, insufficient result stability, and difficulty in balancing loading efficiency and transportation costs in complex delivery scenarios, thus limiting the effectiveness of the algorithms in practical logistics systems. Summary of the Invention

[0004] In view of the problems existing in a multi-objective vehicle routing optimization method that considers both load factor and transportation distance, this invention is proposed. Therefore, the problem to be solved by this invention is how to provide a multi-objective vehicle routing optimization method that considers both load factor and transportation distance.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] In a first aspect, the present invention provides a multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage, comprising: obtaining order data of orders to be delivered from an order management platform, wherein the order data includes order identifier, cargo weight, cargo volume and delivery address; obtaining vehicle rated load and current location data from a vehicle management system; constructing an order load feature vector based on cargo weight and cargo volume; clustering orders according to load differences between orders to form multiple order clusters; calculating the total weight and total volume of each order cluster; and assigning an initial priority label to each order cluster according to the comprehensive load characteristics.

[0007] The order clusters are sorted according to the initial priority labels, and the order clusters are merged based on the spatial location information and loading rate constraints of the order clusters, and the initial vehicle routes are generated.

[0008] A multi-objective collaborative optimization path evaluation model is constructed to optimize the initial vehicle path, generate a set of alternative path schemes, calculate a comprehensive score for the alternative path schemes, and select the path scheme with the highest comprehensive score as the final path scheme output.

[0009] As a preferred embodiment of the multi-objective vehicle routing optimization method that considers both loading rate and transportation mileage as described in this invention, the step of clustering orders based on load differences between orders includes:

[0010] The order load feature vectors are scaled uniformly, and the load difference between any two orders is calculated based on the scaled uniform feature vectors.

[0011] A clustering algorithm is used to merge orders with load differences less than a preset threshold into the same order cluster, generating multiple order clusters.

[0012] As a preferred embodiment of the multi-objective vehicle routing optimization method that considers both loading rate and transportation mileage as described in this invention, the step of assigning initial priority labels to each order cluster based on comprehensive load characteristics includes:

[0013] The comprehensive load characterization index is calculated based on the total weight and total volume of the order cluster. The comprehensive load characterization index of all order clusters is compared to establish the relative load hierarchy relationship between order clusters.

[0014] Priority labels are assigned to order clusters based on the relative load hierarchy, such that the priority labels are positively correlated with the relative load hierarchy.

[0015] As a preferred embodiment of the multi-objective vehicle routing optimization method that considers both loading rate and transportation mileage as described in this invention, the step of merging order clusters based on the spatial location information of order clusters and loading rate constraints includes:

[0016] Extract the spatial location information of each order cluster and convert it into a unified spatial feature representation;

[0017] Calculate the spatial distance metric between any two order clusters, and merge the order clusters to form several order cluster sets, provided that the loading rate constraint is satisfied; the loading rate constraint includes weight loading rate constraint and volume loading rate constraint.

[0018] Each order cluster is grouped into an independent transportation batch. An initial vehicle route is constructed based on the spatial distribution relationship of the order clusters within the transportation batch. The spatial locations of each order cluster within the transportation batch are connected in series according to the principle of distance continuity to form an initial driving route covering all delivery nodes of the transportation batch.

[0019] As a preferred embodiment of the multi-objective vehicle route optimization method that considers both loading rate and transportation mileage as described in this invention, the route evaluation model for multi-objective collaborative optimization includes total driving distance, total driving time, and loading utilization rate, which are defined as follows:

[0020]

[0021]

[0022]

[0023] in, Total distance traveled. For adjacent delivery nodes in the path and Spatial distance between These are the adjacent node pairs visited in the initial path scheme. Total travel time For vehicles from delivery nodes Drive to delivery node The time required To maximize loading utilization, This is the combined weight requirement for all order clusters within the shipping batch. The rated load capacity of the vehicle for executing the route plan.

[0024] As a preferred embodiment of the multi-objective vehicle route optimization method that considers both loading rate and transportation mileage as described in this invention, the generation of the alternative route scheme set includes:

[0025] The overall objective function for constructing the initial path scheme is expressed as:

[0026]

[0027] in, The overall objective function for the initial path scheme is... , and Weighting coefficients for the evaluation objectives;

[0028] With minimizing the comprehensive objective function as the optimization objective, a path optimization algorithm is used to adjust the initial vehicle path, iteratively generating multiple feasible path schemes to form a set of alternative path schemes.

[0029] As a preferred embodiment of the multi-objective vehicle route optimization method that considers both loading rate and transportation mileage as described in this invention, the step of calculating a comprehensive score for alternative route options includes:

[0030] For each alternative route, calculate the index values ​​for total driving distance, total driving time, and load utilization rate;

[0031] The values ​​of each indicator are normalized to obtain a normalized score for each indicator. A weighted summation method is then used to calculate the comprehensive score of the alternative path schemes, expressed as:

[0032]

[0033] in, The overall score of the alternative route options, For alternative route options in the first Normalized scoring results under each evaluation indicator For the first The weighting coefficients of each evaluation indicator;

[0034] The candidate route options are sorted in descending order based on the comprehensive score, and the option with the highest comprehensive score is selected as the final route option.

[0035] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a multi-objective vehicle route optimization method that takes into account both load rate and transportation mileage.

[0036] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements the steps of a multi-objective vehicle route optimization method that takes into account both load rate and transportation mileage.

[0037] The beneficial effects of this invention are as follows: By uniformly modeling order weight and volume as load feature vectors and performing clustering, this invention achieves structured grouping of orders at the load level, effectively reducing scheduling complexity; by combining order cluster priority and spatial proximity, initial vehicle routes are generated under loading rate constraints, avoiding imbalance in vehicle resource allocation; furthermore, a multi-objective collaborative optimization model is used to comprehensively evaluate and screen routes, achieving an overall balance between transportation mileage, loading utilization rate, and route rationality, thereby improving vehicle loading efficiency and delivery efficiency, and enhancing the stability and practical feasibility of route decisions. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of a multi-objective vehicle routing optimization method that takes into account both load factor and transport distance. Detailed Implementation

[0040] To make the above-mentioned objects, features, and advantages of the present invention more readily understood, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0041] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0042] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0043] Reference Figure 1 This is the first embodiment of the present invention, which provides a multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage, including:

[0044] S1: Obtain order data for orders to be delivered from the order management platform. The order data includes order identifier, cargo weight, cargo volume, and delivery address. Obtain vehicle rated load and current location data from the vehicle management system. Construct an order load feature vector based on cargo weight and cargo volume. Cluster the orders according to the load differences between orders to form multiple order clusters. Calculate the total weight and total volume of each order cluster. Assign an initial priority label to each order cluster based on the comprehensive load characteristics.

[0045] S2: Sort the order clusters according to the initial priority labels, merge the order clusters based on the spatial location information and loading rate constraints, and generate the initial vehicle routes;

[0046] S3: Construct a multi-objective collaborative optimization path evaluation model to optimize the initial vehicle path, generate a set of alternative path schemes, calculate a comprehensive score for the alternative path schemes, and select the path scheme with the highest comprehensive score as the final path scheme output.

[0047] Specifically, the order management platform centrally retrieves all orders currently awaiting delivery and obtains each basic data item for each order.

[0048] The basic data items include order identification information to uniquely identify the order, weight and volume information reflecting the physical characteristics of the goods in the order, and delivery address or geographical location description information for subsequent spatial analysis;

[0049] At the same time scale after the order data retrieval is completed, the vehicle operation data that is currently in a dispatchable state is synchronously obtained from the vehicle management system. The vehicle operation data includes at least the vehicle's rated load capacity and the vehicle's current location description information.

[0050] The weight and volume information of the orders are processed to construct an order load feature vector for each order;

[0051] The order load feature vectors are scaled uniformly. Based on the scale-uniformed set of order load feature vectors, a similarity metric is constructed between orders to represent the degree of similarity in load demand characteristics between different orders and the load differences between orders. Represented as:

[0052]

[0053] in, For the first The weight of the goods in each order. For the first The volume of goods in each order. For the first The weight of the goods in each order. For the first The volume of goods in each order; the load difference between orders is used to reflect the distance between orders in the joint feature space of weight and volume. The smaller the distance value, the closer the two orders are in terms of load requirement characteristics.

[0054] Clustering is performed on the order set to group orders with similar load characteristics into the same order cluster, so that the orders within the same order cluster are consistent in terms of overall weight and volume, while different order clusters are clearly distinguished in terms of load requirements.

[0055] For each order cluster, the comprehensive load characteristics of the internal orders are statistically analyzed to form an order cluster aggregation description. The total weight and total volume of the order cluster are expressed as follows:

[0056]

[0057]

[0058] in, For the first The total weight of each order cluster For the first The total volume of each order cluster For the first Order clusters.

[0059] A comprehensive load characterization index for order clusters is constructed based on the total weight and total volume of the order clusters. This index reflects the degree to which order clusters occupy vehicle resources. The comprehensive load characterization indices of all order clusters are compared to form a relative load hierarchy relationship between order clusters.

[0060] Assign an initial priority label to each order cluster. The initial priority label is used to characterize the order of processing of the order cluster in the path planning process.

[0061] The initial priority label allocation results correspond one-to-one with the comprehensive load characteristics of the order cluster, so that the order clusters with large load scale and high resource consumption have clear scheduling order identifiers in the overall delivery process, thus providing a stable and repeatable input basis for batch merging and path optimization based on vehicle capacity constraints in subsequent steps.

[0062] All order clusters are globally sorted according to the initial priority label. After sorting the order clusters, spatial location information is extracted for each order cluster and converted into a unified spatial feature representation to represent the relative positional relationship of order clusters in geographical distribution. A spatial distance metric is constructed for any two order clusters.

[0063] When determining the order cluster merging, a loading rate constraint is introduced. For any candidate order cluster combination, the total weight and total volume of the combination are obtained.

[0064] The combined load characteristics are matched with the rated load capacity of the target vehicle to calculate the weight loading rate and volume loading rate. Under the premise of meeting the loading rate constraint, the spatial proximity between order clusters is used as an important basis for merging decision. By comparing spatial distance metrics, order clusters that are geographically adjacent and have feasible loading combinations are selected for merging first.

[0065] The merging operation of order clusters is carried out in a stepwise expansion manner. After each merging is completed, the combined load characteristics and spatial location characteristics of the current merging result are updated.

[0066] Several order clusters are formed, and each order cluster is an independent transportation batch. An initial vehicle path is constructed based on the spatial distribution relationship of the order clusters within the transportation batch. The spatial positions of each order cluster within the transportation batch are connected in series according to the principle of distance continuity to form an initial driving path covering all delivery nodes of the transportation batch, which is used to represent the basic driving sequence of the vehicles.

[0067] After obtaining the transportation batch delivery task and the initial vehicle route, the batch delivery task and the initial vehicle route are modeled in a unified manner. The set of order clusters contained in the transportation batch, the spatial location information of the order clusters, and the access order of each delivery node in the initial vehicle route are integrated to form a joint description object for route optimization.

[0068] Based on the joint description object, a multi-objective collaborative optimization path evaluation model is constructed, which incorporates multiple objectives involved in the path planning process into a unified optimization framework. For any path scheme, the objectives of total travel distance, total travel time, and load utilization rate are defined as follows:

[0069]

[0070]

[0071]

[0072] in, Total distance traveled. For adjacent delivery nodes in the path and Spatial distance between These are the adjacent node pairs visited in the initial path scheme. Total travel time For vehicles from delivery nodes Drive to delivery node The time required To maximize loading utilization, This is the combined weight requirement for all order clusters within the shipping batch. The rated load capacity of the vehicle for executing the route plan;

[0073] Based on multiple evaluation objectives, a comprehensive objective function for the initial path scheme is constructed, expressed as:

[0074]

[0075] in, The overall objective function for the initial path scheme is... , and Weighting coefficients for the evaluation objectives;

[0076] By minimizing the comprehensive objective function, a path optimization algorithm is used to adjust the initial vehicle path, gradually generating multiple feasible path schemes under different objective trade-offs, which are then collected to form a set of alternative path schemes; each alternative path scheme corresponds to a complete set of delivery node sequences and vehicle driving paths.

[0077] The initial vehicle route is used as the input object of the route optimization algorithm. The set of delivery nodes in the route and their access order are analyzed to clarify the starting point, ending point and the sequential relationship between each delivery node. The load constraints and vehicle capacity constraints of the corresponding transportation batch are also included in the set of constraints for route adjustment.

[0078] While keeping the path constraints unchanged, the access order of delivery nodes in the path is gradually rearranged, and new candidate paths are generated through local adjustments. During multiple path adjustments, the path cost of each generated candidate path is calculated and compared with the current path. When the candidate path is better than the current path in the overall evaluation result, the current path is updated as the new optimal path.

[0079] Once the route adjustment process reaches the preset termination condition, the optimized vehicle route is output and collected to form a set of alternative route schemes.

[0080] After generating the set of alternative route options, each alternative route option is evaluated uniformly. The calculation results of each evaluation index are normalized to obtain the comprehensive score of the alternative route options, expressed as:

[0081]

[0082] in, The overall score of the alternative route options, For alternative route options in the first Normalized scoring results under each evaluation indicator For the first The weighting coefficients of each evaluation indicator;

[0083] Based on the comprehensive scoring results, the set of alternative route options is sorted, and the route option with the highest score is selected as the final route option output, thereby providing a stable, repeatable and efficient route decision-making basis for actual delivery execution.

[0084] This embodiment also provides a computer device applicable to a multi-objective vehicle route optimization method that takes into account both load rate and transportation mileage, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement all or part of the steps of the method described in the above embodiments of the present invention.

[0085] This embodiment also provides a storage medium storing a computer program thereon. When the computer program is executed by a processor, it performs the method in any optional implementation of the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0086] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0087] In summary, this invention achieves structured grouping of orders at the load level by uniformly modeling order weight and volume as load feature vectors and performing clustering, effectively reducing scheduling complexity. By combining order cluster priority and spatial proximity, initial vehicle routes are generated under loading rate constraints, avoiding imbalances in vehicle resource allocation. Furthermore, a multi-objective collaborative optimization model comprehensively evaluates and filters routes, achieving an overall balance between transportation mileage, loading utilization, and route rationality, thereby improving vehicle loading efficiency and delivery efficiency, and enhancing the stability and practical feasibility of route decisions.

[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-objective vehicle route optimization method that considers both loading rate and transportation distance, characterized in that: include, Order data for orders to be delivered is obtained from the order management platform. The order data includes order identifier, cargo weight, cargo volume, and delivery address. The rated load and current location data of the vehicles are obtained from the vehicle management system. An order load feature vector is constructed based on the cargo weight and cargo volume. The orders are clustered according to the load differences between orders to form multiple order clusters. The total weight and total volume of each order cluster are calculated. An initial priority label is assigned to each order cluster according to the comprehensive load characteristics. The order clusters are sorted according to the initial priority labels, and the order clusters are merged based on the spatial location information and loading rate constraints of the order clusters, and the initial vehicle routes are generated. A multi-objective collaborative optimization path evaluation model is constructed to optimize the initial vehicle path, generate a set of alternative path schemes, calculate a comprehensive score for the alternative path schemes, and select the path scheme with the highest comprehensive score as the final path scheme output.

2. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 1, characterized in that: The clustering of orders based on load differences between orders includes: The order load feature vectors are scaled uniformly, and the load difference between any two orders is calculated based on the scaled uniform feature vectors. A clustering algorithm is used to merge orders with load differences less than a preset threshold into the same order cluster, generating multiple order clusters.

3. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 1, characterized in that: The process of assigning an initial priority label to each order cluster based on comprehensive load characteristics includes: The comprehensive load characterization index is calculated based on the total weight and total volume of the order cluster. The comprehensive load characterization index of all order clusters is compared to establish the relative load hierarchy relationship between order clusters. Priority labels are assigned to order clusters based on the relative load hierarchy, such that the priority labels are positively correlated with the relative load hierarchy.

4. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 1, characterized in that: The process of merging order clusters based on their spatial location information and loading rate constraints includes: Extract the spatial location information of each order cluster and convert it into a unified spatial feature representation; Calculate the spatial distance metric between any two order clusters, and merge the order clusters to form several order cluster sets, provided that the loading rate constraint is satisfied; the loading rate constraint includes weight loading rate constraint and volume loading rate constraint. Each order cluster is grouped into an independent transportation batch. An initial vehicle route is constructed based on the spatial distribution relationship of the order clusters within the transportation batch. The spatial locations of each order cluster within the transportation batch are connected in series according to the principle of distance continuity to form an initial driving route covering all delivery nodes of the transportation batch.

5. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 4, characterized in that: The multi-objective collaborative optimization path evaluation model includes total travel distance, total travel time, and load utilization rate, which are defined as follows: ； ； ； in, Total distance traveled. For adjacent delivery nodes in the path and Spatial distance between These are the adjacent node pairs visited in the initial path scheme. Total travel time For vehicles from delivery nodes Drive to delivery node The time required To maximize loading utilization, This is the combined weight requirement for all order clusters within the shipping batch. The rated load capacity of the vehicle for executing the route plan.

6. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 1, characterized in that: The generated set of alternative path schemes includes: The overall objective function for constructing the initial path scheme is expressed as: ； in, The overall objective function for the initial path scheme is... , and Weighting coefficients for the evaluation objectives; With minimizing the comprehensive objective function as the optimization objective, a path optimization algorithm is used to adjust the initial vehicle path, iteratively generating multiple feasible path schemes to form a set of alternative path schemes.

7. The multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in claim 6, characterized in that: The comprehensive scoring calculation for the alternative route schemes includes: For each alternative route, calculate the index values ​​for total driving distance, total driving time, and load utilization rate; The values ​​of each indicator are normalized to obtain a normalized score for each indicator. A weighted summation method is then used to calculate the comprehensive score of the alternative path schemes, expressed as: ； in, The overall score of the alternative route options, For alternative route options in the first Normalized scoring results under each evaluation indicator For the first The weighting coefficients of each evaluation indicator; The candidate route options are sorted in descending order based on the comprehensive score, and the option with the highest comprehensive score is selected as the final route option.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the multi-objective vehicle route optimization method that takes into account both loading rate and transportation mileage as described in any one of claims 1 to 7.

Images