Heuristic algorithm-based path optimization method and system

Abstract

The invention discloses a path optimization method and system based on a heuristic algorithm, and belongs to the technical field of logistics distribution supply. The method comprises the following steps of: constructing a hierarchical network: constructing a hierarchical membership network of large-area nodes, middle-area nodes and cell nodes in a manner of combining an initial distribution node membership and a global optimization node hierarchical relationship; and hierarchical constraint path planning: based on the constructed hierarchical membership network, generating a distribution path following a flow direction from a large region level to a middle region level to a small region level for each large region node, synchronously optimizing a path access sequence and transportation mode selection, and realizing collaborative optimization of timeliness and economy. The invention provides a material distribution optimization scheme which adapts to multi-region hierarchical requirements, gives consideration to multi-efficiency targets and ensures path feasibility, finally realizes accurate, efficient and economical delivery of important materials, and provides technical support for construction of a modern logistics system.

Description

Technical Field

[0001] This invention belongs to the field of logistics distribution and replenishment technology, specifically involving a path optimization method and system based on heuristic algorithms, which is used to achieve multi-objective collaborative optimization of material distribution and improve delivery timeliness and economy. Background Technology

[0002] In modern society, the timeliness and economy of goods distribution have become core competitiveness and the "lifeline" of economic and social operation. Timeliness is the lifeblood of modern commerce. In the fast-paced consumption era, "instant gratification" and "on-demand delivery" have become the norm. Rapid delivery is not only key to improving user experience and enhancing customer loyalty, but also a direct reflection of the agility of a company's supply chain. An efficient logistics network ensures the continuity and stability of social production and life, greatly reducing the waiting costs and inventory costs for the entire society. Economy is the cornerstone of market expansion. Low distribution costs are an important component of commodity price competitiveness, making cross-regional, large-scale goods exchange possible. For suppliers, optimizing logistics routes and reducing transportation expenses are core elements for controlling overall operating costs, increasing profit margins, and enhancing sustainable competitiveness. The production, storage, and transportation of important materials involve huge resource inputs and complex supply chains. Pursuing the economy of distribution while meeting usage needs means achieving maximum protection benefits with minimal resource consumption and the lowest logistics costs.

[0003] Existing technologies for optimizing material distribution routes have significant shortcomings: First, most route optimization methods focus on single-level distribution scenarios, failing to coordinate the hierarchical relationships between multiple nodes and thus failing to meet the needs of hierarchical material delivery. Second, the optimization objectives are singular, either emphasizing only time efficiency or cost control, without dynamically balancing the efficiency priorities at different stages, and failing to consider the synergy between timeliness and economy. Third, there is a lack of adaptive design for distribution needs in complex environments, resulting in insufficient route feasibility and robustness, making it difficult to support precise delivery.

[0004] The aforementioned shortcomings make it difficult for existing technologies to meet the actual needs of modern society for the efficient and stable delivery of essential materials. Summary of the Invention

[0005] The technical problem to be solved by this invention is to improve the timeliness and economy of cross-regional material distribution. It provides a path optimization method and system based on heuristic algorithms, and designs an adapted distribution network for a three-level system of large, medium and small regions. It solves specific problems such as the coordinated optimization of timeliness and economy under rigid hierarchical constraints, the global planning of multi-level node affiliation, and the guarantee of path feasibility in complex environments, so as to meet the actual needs of accurate material delivery.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A path optimization method based on heuristic algorithms includes the following steps: Step S1, Hierarchical Network Construction: By combining the initial allocation of node affiliation relationships with the global optimization of node hierarchy relationships, a hierarchical affiliation network of regional nodes, medium-region nodes, and cell nodes is constructed. Step S2, Hierarchical Constraint Path Planning: Based on the hierarchical membership network constructed in Step S1, a delivery path following the flow direction of regional level → intermediate level → sub-regional level is generated for each regional node. The path access sequence and transportation mode selection are optimized simultaneously to achieve synergistic optimization of timeliness and economy.

[0007] Furthermore, in step S1, the construction of the hierarchical network specifically includes: S1.1, Initial affiliation allocation: Assign each central area node to the nearest large area node, and assign each cell node to the nearest central area node, thus forming the initial hierarchical affiliation; S1.2, Global Optimization of Hierarchical Relationships: The hierarchical relationships of nodes are represented by a double-integer array encoding, where the first array represents the allocation relationship from central area nodes to large area nodes, and the second array represents the allocation relationship from small area nodes to central area nodes; the total transportation cost is used as the optimization objective, where the total transportation cost is α × total transportation time + β × total transportation cost, and α and β are dynamically adjusted weight coefficients; combined with tournament selection, multi-point crossover, and constraint repair mutation operations, an optimal hierarchical network is generated.

[0008] Furthermore, in step S1.2, the constraint repair mutation operation specifically involves: when an illegal state of a node being unassigned occurs after crossover or mutation, the unassigned node is reassigned to the nearest legal parent node.

[0009] Furthermore, in step S2, the hierarchical constraint path planning specifically includes: S2.1, Multi-path generation: Generate multiple independent delivery routes for each regional node. Each route follows the hierarchical access order of regional node → central node → sub-regional node under the central node, and cross-level or reverse access is prohibited. S2.2, Path Coding and Cost Calculation: Variable-length sequence coding is used to represent the path structure, with the coding format being [Large Area ID, Middle Area ID, Cell ID sequence]; the path cost is calculated segment by segment, where the cost of the Large Area → Middle Area segment is γ × Time Cost + δ × Cost Cost, and the cost of the Middle Area → Cell segment is γ × Time Cost + δ × Cost Cost. Furthermore, the weight of γ for the Large Area → Middle Area segment is higher than that for the Middle Area → Cell segment, and the weight of δ for the Middle Area → Cell segment is higher than that for the Large Area → Middle segment. S2.3, Dynamic Optimization: For the cluster of community nodes, when the number of nodes is ≤8, the optimal sequence is selected by traversing all possible access sequences; when the number of nodes is >8, the access sequence is optimized by heuristic search. At the same time, the transportation mode is dynamically selected for each path segment, and the transportation mode includes road, rail and air transport.

[0010] Furthermore, step S2.1 also includes an illegal path penalty mechanism: if a path involves cross-level access or does not fully cover its own node, a high penalty is imposed on the path.

[0011] Furthermore, the selection of the transportation mode is achieved through extended coding, which adds a transportation mode gene bit to the path coding, where 0 represents road, 1 represents rail, and 2 represents air transport. The transportation mode is dynamically adjusted according to real-time road conditions and cost weights.

[0012] A path optimization system based on a heuristic algorithm includes a connected hierarchical network construction module and a hierarchical constraint path planning module. The hierarchical network construction module is used to construct a hierarchical membership network of large-area nodes, medium-area nodes, and small-area nodes by combining initial allocation of node membership relationships with global optimization of node hierarchical relationships. The hierarchical constraint path planning module is used to generate a delivery path for each large-area node based on the hierarchical membership network constructed by the hierarchical network construction module, following the flow direction of large-area level → medium-area level → small-area level, and simultaneously optimize the path access sequence and transportation mode selection.

[0013] Furthermore, the hierarchical network construction module includes an initial membership assignment submodule and a hierarchical relationship global optimization submodule. The initial membership assignment submodule is used to assign each central node to the nearest large node and each small node to the nearest central node, forming an initial hierarchical membership relationship. The hierarchical relationship global optimization submodule is used to represent the node hierarchical relationship through a double-integer array encoding, with the total transportation cost as the optimization objective, and combines tournament selection, multi-point crossover, and constraint repair mutation operations to generate an optimal hierarchical network. The total transportation cost is α × total transportation time + β × total transportation cost, where α and β are dynamically adjusted weight coefficients.

[0014] Furthermore, the hierarchical relationship global optimization submodule is also used to: when an illegal state of a node having no affiliation occurs after crossover or mutation, the node without affiliation is reassigned to the nearest legal parent node.

[0015] Furthermore, the hierarchical constraint path planning module includes a path generation submodule, an encoding and cost calculation submodule, and a dynamic optimization submodule. The path generation submodule generates multiple independent delivery paths for each regional node. Each path follows a hierarchical access order of regional node → central regional node → sub-regional node under the central regional node, prohibiting cross-level or reverse access. The encoding and cost calculation submodule uses variable-length sequence encoding to represent the path structure and calculates the path cost in segments. The regional → central regional segment emphasizes timeliness weight γ, and the central regional → sub-regional segment emphasizes economic weight δ. The dynamic optimization submodule optimizes the access sequence for the sub-regional node cluster by traversing all possible access sequences or using heuristic search, and dynamically selects the transportation mode of road, rail, or air for each path segment.

[0016] The beneficial effects of this invention are: This invention constructs a three-tiered distribution network—regional, intermediate, and sub-regional—clearly defining node affiliations and rigid flow constraints. This allows for precise matching with modern cross-regional distribution systems, preventing cross-level distribution chaos and ensuring the orderly transfer of goods across different levels. It solves the problem of existing technologies' inability to coordinate multi-level nodes, improving the targeting and economy of distribution. Addressing the different needs at different stages and levels, dynamic weighting design—emphasizing timeliness in emergencies and economy in long-term needs—achieves multi-objective collaborative optimization. This ensures timely response to critical material replenishment while reducing waste of logistics resources, such as lowering empty load rates and optimizing capacity turnover, achieving optimal resource allocation and overcoming the shortcomings of single-objective optimization in existing technologies. Through designs such as high penalties for illegal routes, automatic repair of unaffiliated nodes, and dynamic adaptation to multiple transportation modes, it can cope with path uncertainties caused by complex environments, ensuring the feasibility and robustness of distribution routes under strong constraints and supporting continuous distribution under high-intensity demand.

[0017] The multi-level collaborative distribution optimization mechanism formed by this invention can be directly integrated into logistics systems in various fields, realizing agile and intensive management of the distribution of important materials, providing core technical support for building an on-demand material distribution system, and helping to improve overall support efficiency. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system structure diagram of the present invention. Detailed Implementation

[0019] To facilitate understanding of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Those skilled in the art should understand that the embodiments described are merely illustrative of the invention and should not be considered as specific limitations thereof.

[0020] The three-tiered system of large region, medium region, and small region mentioned in this invention is mainly divided according to geographical area, referring to the national region, the main directional region, and the provincial and lower-level regions, respectively. Example 1

[0021] like Figure 1 As shown, a path optimization method based on heuristic algorithms includes the following steps: Step S1, Hierarchical Network Construction: By combining the initial allocation of node affiliation relationships with the global optimization of node hierarchy relationships, a hierarchical affiliation network of regional nodes, medium-region nodes, and small-region nodes is constructed.

[0022] This step uses a combination of greedy and genetic algorithms to construct a hierarchical membership network, specifically including: S1.1, Initial affiliation allocation: Assign each central area node to the nearest large area node, and assign each cell node to the nearest central area node, thus forming the initial hierarchical affiliation; S1.2, Global Optimization of Hierarchical Relationships: The hierarchical relationships of nodes are represented by a double-integer array encoding, where the first array represents the allocation relationship from central area nodes to large area nodes, and the second array represents the allocation relationship from small area nodes to central area nodes; the total transportation cost is used as the optimization objective, which is α × total transportation time + β × total transportation cost, where α and β are weight coefficients dynamically adjusted according to logistics priority. For example, when facing urgent needs, α > β emphasizes timeliness; the optimal hierarchical network is generated by combining tournament selection, multi-point crossover, and constraint repair mutation operations.

[0023] The constraint repair mutation operation is as follows: when an illegal state of no-home node occurs after crossover or mutation, the no-home node is reassigned to the nearest legal parent node.

[0024] Step S2, Hierarchical Constraint Path Planning: Based on the hierarchical membership network constructed in Step S1, a delivery path following the flow direction of regional level → intermediate level → sub-regional level is generated for each regional node. The path access sequence and transportation mode selection are optimized simultaneously to achieve synergistic optimization of timeliness and economy.

[0025] Hierarchical constraint path planning specifically includes: S2.1, Multi-path generation: Generate multiple independent delivery routes for each regional node. Each route follows the hierarchical access order of regional node → central node → sub-regional node under the central node, and cross-level or reverse access is prohibited. This step also includes an illegal path penalty mechanism: if a path involves cross-level access or does not fully cover its own node, a high penalty will be imposed on the path, with a penalty value ≥ 10. 6 .

[0026] S2.2, Route Coding and Cost Calculation: Variable-length sequence coding is used to represent the route structure, with the coding format being [Large Area ID, Medium Area ID, Small Area ID sequence]. The route cost is calculated segment by segment, where the cost of the Large Area → Medium Area segment is γ × Time Cost + δ × Cost, and the cost of the Medium Area → Small Area segment is γ × Time Cost + δ × Cost. The γ weight of the Large Area → Medium Area segment (long-distance trunk line) is higher than that of the Medium Area → Small Area segment, emphasizing timeliness, while the δ weight of the Medium Area → Small Area segment (last-mile delivery) is higher than that of the Large Area → Medium Area segment, emphasizing economy. S2.3, Dynamic Optimization: For the cluster of community nodes, when the number of nodes is ≤8, the optimal sequence is selected by traversing all possible access sequences. When the number of nodes is >8, the access sequence is optimized by heuristic search. At the same time, the transportation mode is dynamically selected for each path segment. The transportation modes include road, rail and air.

[0027] The selection of transportation mode is achieved through extended coding, which adds a transportation mode gene bit to the route code, where 0 represents road, 1 represents rail, and 2 represents air transport. The transportation mode is dynamically adjusted according to real-time road conditions and cost weights.

[0028] This invention achieves multi-path collaborative optimization by using gene-encoded forced hierarchical flow, segmented dynamic weighting, and adaptive selection of transportation modes, while meeting the hard constraints of delivery requirements. This improves timeliness by 15%–22% and reduces transportation costs by 8%–12% compared to traditional methods. The path optimization stage integrates transportation mode selection and node sequence scheduling, enabling dynamic resource allocation. These two aspects work together to ensure the timeliness, economy, and robustness of the material distribution system. Example 2

[0029] like Figure 2 As shown, a path optimization system based on a heuristic algorithm includes a connected hierarchical network construction module and a hierarchical constraint path planning module. The hierarchical network construction module is used to construct a hierarchical membership network of large area nodes, medium area nodes, and small area nodes by combining the initial allocation of node membership relationships with the global optimization of node hierarchical relationships. The hierarchical constraint path planning module is used to generate a delivery path for each large area node based on the hierarchical membership network constructed by the hierarchical network construction module, following the flow direction of large area level → medium area level → small area level, and simultaneously optimize the path access sequence and transportation mode selection.

[0030] The hierarchical network construction module includes an initial membership assignment submodule and a hierarchical relationship global optimization submodule. The initial membership assignment submodule assigns each central node to the nearest large node and each small node to the nearest central node, forming the initial hierarchical membership. The hierarchical relationship global optimization submodule uses a double-integer array to encode the node hierarchical relationship, takes the total transportation cost as the optimization objective, and combines tournament selection, multi-point crossover, and constraint repair mutation operations to generate the optimal hierarchical network. The total transportation cost is α × total transportation time + β × total transportation cost, where α and β are dynamically adjusted weight coefficients. When an illegal state of no-home node occurs after crossover or mutation, the no-home node is reassigned to the nearest legal parent node.

[0031] The hierarchical constraint path planning module includes a path generation submodule, an encoding and cost calculation submodule, and a dynamic optimization submodule. The path generation submodule generates multiple independent delivery paths for each regional node. Each path follows the hierarchical access order of regional node → central node → sub-regional node under the central node, prohibiting cross-level or reverse access. The encoding and cost calculation submodule uses variable-length sequence encoding to represent the path structure and calculates the path cost in segments. The regional → central segment emphasizes timeliness weight γ, and the central → sub-regional segment emphasizes economic weight δ. The dynamic optimization submodule optimizes the access sequence for the sub-regional node cluster by traversing all possible access sequences (number of nodes ≤ 8) or by heuristic search (number of nodes > 8), and dynamically selects the transportation mode of road, rail, or air for each path segment.

[0032] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A path optimization method based on heuristic algorithms, characterized in that, Includes the following steps: Step S1, Hierarchical Network Construction: By combining the initial allocation of node affiliation relationships with the global optimization of node hierarchy relationships, a hierarchical affiliation network of regional nodes, medium-region nodes, and cell nodes is constructed. Step S2, Hierarchical Constraint Path Planning: Based on the hierarchical membership network constructed in Step S1, a delivery path following the flow direction of regional level → intermediate level → sub-regional level is generated for each regional node. The path access sequence and transportation mode selection are optimized simultaneously to achieve synergistic optimization of timeliness and economy.

2. The path optimization method based on heuristic algorithms according to claim 1, characterized in that, In step S1, the construction of the hierarchical network specifically includes: S1.1, Initial affiliation allocation: Assign each central area node to the nearest large area node, and assign each cell node to the nearest central area node, thus forming the initial hierarchical affiliation; S1.2, Global Optimization of Hierarchical Relationships: The hierarchical relationships of nodes are represented by a double-integer array encoding, where the first array represents the allocation relationship from central area nodes to large area nodes, and the second array represents the allocation relationship from small area nodes to central area nodes; the total transportation cost is used as the optimization objective, where the total transportation cost is α × total transportation time + β × total transportation cost, and α and β are dynamically adjusted weight coefficients; combined with tournament selection, multi-point crossover, and constraint repair mutation operations, an optimal hierarchical network is generated.

3. The path optimization method based on heuristic algorithms according to claim 2, characterized in that, In step S1.2, the constraint repair mutation operation specifically means that when an illegal state of no-home node occurs after crossover or mutation, the no-home node is reassigned to the nearest legal parent node.

4. The path optimization method based on heuristic algorithms according to claim 1, characterized in that, In step S2, the hierarchical constraint path planning specifically includes: S2.1, Multi-path generation: Generate multiple independent delivery routes for each regional node. Each route follows the hierarchical access order of regional node → central node → sub-regional node under the central node, and cross-level or reverse access is prohibited. S2.2, Path Coding and Cost Calculation: Variable-length sequence coding is used to represent the path structure, with the coding format being [Large Area ID, Middle Area ID, Cell ID sequence]; the path cost is calculated segment by segment, where the cost of the Large Area → Middle Area segment is γ × Time Cost + δ × Cost Cost, and the cost of the Middle Area → Cell segment is γ × Time Cost + δ × Cost Cost. Furthermore, the weight of γ for the Large Area → Middle Area segment is higher than that for the Middle Area → Cell segment, and the weight of δ for the Middle Area → Cell segment is higher than that for the Large Area → Middle segment. S2.3, Dynamic Optimization: For the cluster of community nodes, when the number of nodes is ≤8, the optimal sequence is selected by traversing all possible access sequences; when the number of nodes is >8, the access sequence is optimized by heuristic search. At the same time, the transportation mode is dynamically selected for each path segment, and the transportation mode includes road, rail and air transport.

5. The path optimization method based on heuristic algorithms according to claim 4, characterized in that, Step S2.1 also includes an illegal path penalty mechanism: if a path involves cross-level access or does not fully cover its own node, a high penalty is imposed on the path.

6. The path optimization method based on heuristic algorithms according to claim 4, characterized in that, In step S2.3, the selection of the mode of transport is achieved through extended coding, which adds a mode of transport gene bit to the path coding, where 0 represents road, 1 represents rail, and 2 represents air transport. The mode of transport is dynamically adjusted according to real-time road conditions and cost weights.

7. A path optimization system based on a heuristic algorithm, characterized in that, It includes a connected hierarchical network construction module and a hierarchical constraint path planning module; the hierarchical network construction module is used to construct a hierarchical membership network of large area nodes, medium area nodes and small area nodes by combining the initial allocation of node membership relationships with the global optimization of node hierarchical relationships; The hierarchical constraint path planning module is used to generate a delivery path for each regional node based on the hierarchical membership network constructed by the hierarchical network construction module, following the flow direction of regional level → intermediate level → sub-region level, and simultaneously optimizes the path access sequence and transportation mode selection.

8. The path optimization system based on heuristic algorithm according to claim 7, characterized in that, The hierarchical network construction module includes an initial membership assignment submodule and a hierarchical relationship global optimization submodule; the initial membership assignment submodule is used to assign each central area node to the nearest large area node, and each cell node to the nearest central area node, forming an initial hierarchical membership relationship; The hierarchical relationship global optimization submodule is used to represent the node hierarchical relationship through double integer array encoding. With the total transportation cost as the optimization objective, it combines tournament selection, multi-point crossover, and constraint repair mutation operations to generate the optimal hierarchical network. The total transportation cost is α × total transportation time + β × total transportation cost, where α and β are dynamically adjusted weight coefficients.

9. The path optimization system based on heuristic algorithms according to claim 8, characterized in that, The hierarchical relationship global optimization submodule is also used to: when an illegal state of a node being unassigned occurs after crossover or mutation, the unassigned node is reassigned to the nearest legal parent node.

10. The path optimization system based on heuristic algorithm according to claim 7, characterized in that, The hierarchical constraint path planning module includes a path generation submodule, an encoding and cost calculation submodule, and a dynamic optimization submodule. The path generation submodule generates multiple independent delivery paths for each regional node. Each path follows a hierarchical access order of regional node → central regional node → sub-regional node under the central regional node, prohibiting cross-level or reverse access. The encoding and cost calculation submodule uses variable-length sequence encoding to represent the path structure and calculates the path cost in segments. The regional → central regional segment emphasizes timeliness weight γ, and the central regional → sub-regional segment emphasizes economic weight δ. The dynamic optimization submodule optimizes the access sequence for the sub-regional node cluster by traversing all possible access sequences or using heuristic search, and dynamically selects the transportation mode (road, rail, or air) for each path segment.

Images