Dynamic logistics network path optimization method based on chain state routing algorithm

Abstract

The invention discloses a dynamic logistics network path optimization method based on a chained state routing algorithm, and belongs to the technical field of logistics network optimization, and the method comprises the steps: firstly, building a node state evaluation model through a distributed state collection mechanism based on a chained state routing protocol, and generating a dynamic elastic network topology structure; secondly, when logistics network nodes change, a logistics path is optimized in combination with a transportation cost prediction model, an emergency response punishment mechanism and a path adjustment weight coefficient; and finally, optimizing an affected path in a node abnormal state by using a local path updating mechanism. Compared with a traditional logistics model, the method has obvious advantages in the aspects of path calculation cost and response time when processing a path problem containing failure nodes, and can effectively optimize logistics path planning, reduce transportation cost and improve system robustness and resource allocation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics network optimization, and particularly to a dynamic logistics network path optimization method based on a chain state routing algorithm. Background Art

[0002] With the rapid development of the logistics industry, logistics path optimization has become one of the key technologies to improve logistics efficiency and reduce transportation costs. Traditional algorithms such as Dijkstra's algorithm and Bellman - Ford's algorithm have significant drawbacks when faced with a dynamically changing logistics network. First, traditional algorithms need to recalculate the path from the source node, and this computational overhead is extremely large in a large - scale network. Especially when a node fails or the network topology changes, recalculating the global path will lead to performance bottlenecks. Second, these algorithms lack a fast response mechanism when dealing with emergencies (such as traffic congestion, node failure), and it is difficult to adjust the path in a timely manner, resulting in a slow - reacting system, increasing logistics costs and delays.

[0003] In addition, traditional path planning algorithms also have limitations when facing a dynamic environment. They usually assume that the network topology is static, lack the ability to adapt to real - time changes, and cannot quickly update path selection. Moreover, existing algorithms fail to fully consider the refined scheduling of logistics resources, resulting in waste of resources and low scheduling efficiency. Therefore, in today's complex and dynamic logistics environment, existing path planning algorithms face significant performance bottlenecks, and there is an urgent need for a new optimization solution to overcome these problems. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention discloses a dynamic logistics network path optimization method based on a chain state routing algorithm to solve the problems raised in the above - mentioned background art.

[0005] To achieve the above object, the present invention provides the following technical solution: A dynamic logistics network path optimization method based on a chain state routing algorithm, characterized by including the following steps:

[0006] S1. Based on the chain state routing protocol, a distributed state acquisition mechanism is used to monitor the operating state of logistics nodes in real - time, build a node state evaluation model, and generate a dynamic elastic network topology structure;

[0007] S2. When changes occur in the logistics network nodes, a dynamic chain routing algorithm is adopted, and the logistics path is optimized by combining a transportation cost prediction model, an emergency response penalty mechanism, and a path adjustment weight coefficient;

[0008] S3. A local path update mechanism is used to optimize and adjust the affected paths under abnormal node states.

[0009] Preferably, in step S1, the operating state of the logistics node includes parameters such as the load level and the processing time limit.

[0010] Preferably, in step S1, the construction of the logistics node state evaluation model includes the following steps:

[0011] S11. Use a piecewise function to judge the node state, and judge according to the working state of the node, such as whether it is faulty, and the conditions of the load situation; define a piecewise function through the health state of the node and the information of the current load, so that when the node is valid, it is 1, and when the node is invalid, it is 0. Judge the validity of node v through the following piecewise function:

[0012]

[0013] Wherein, when the state index x v (t) of node v at time t is greater than or equal to the threshold θ, the value of the function I(x v (t)) is 1. In this situation, it means that node v is valid at this time; when the state index x v (t) of node v at time t is less than the threshold θ, the value of the function I(x v (t)) is 0; that is, node v is invalid at this time;

[0014] S12. Judge whether node v is valid at time t by constructing a mathematical formula based on various state information. The formula form is as follows:

[0015]

[0016] Where L v (t) represents the load of the node at time t, and L max represents the maximum load capacity of the node or path; T v (t) represents the time for the warehousing node to process goods, that is, the time required for the logistics vehicle to enter and leave the warehouse; T max represents the maximum allowed processing time;

[0017] S13. Define T v (t) as the processing time of the node warehouse, and its expression form is as follows:

[0018]

[0019] Where, N w (t) represents the number of workers in the warehouse at time t, N c (t) represents the number of vehicles being processed simultaneously at time t, C w (t) represents the processing capacity of the warehouse, that is, the maximum number of vehicles processed per unit time; F(N c(t)) is the vehicle processing function, representing the number of vehicles N processed by the warehouse at time t c (t)'s impact on the processing time;

[0020] S14. Define F(N c (t)) as the vehicle processing function, representing the number of vehicles N processed by the warehouse at time t c (t)'s impact on the processing time, and its expression form is as follows:

[0021] F(N c (t)) = a·N c (t) b

[0022] where a and b are constants, representing the non - linear impact of the vehicle processing quantity on time. When the number of vehicles increases, the speed of the processing time increases;

[0023] S15. The number of workers and the warehouse processing capacity function G(N w (t), C w (t)), which represents the impact of the number of workers and the warehouse processing capacity on the processing time; as the number of workers N w (t) increases or the warehouse processing capacity C w (t) improves, the processing time T v (t) will decrease, expressed as:

[0024] G(N w (t), C w (t)) = δ·N w (t)+ε·C w (t)

[0025] where δ and ε are constants, reflecting the contributions of the number of workers and the warehouse processing capacity.

[0026] Preferably, in step S2, the path - weight transportation cost prediction model, the emergency response penalty mechanism, and the path - adjustment weight coefficient include the following steps:

[0027] S21. Construct a transportation cost prediction model C(P) to optimize the entire transportation link, and the cost function form is as follows:

[0028]

[0029] where P is the set of paths from the source point s to the target point t, α is a parameter for weighing the normal transportation cost and the emergency - handling cost, f(x v (t), L) is a penalty function used to optimize the path selection to avoid overloaded or failed nodes; L represents the load level in the logistics activity, w ij (t) represents the path weight, i.e., the cost, at time t, expressed as:

[0030] w ij w(t) = d ij + c ij (t)

[0031] where d ij represents the distance or basic transportation cost from node i to node j; c ij (t) represents the dynamic adjustment cost at time t;

[0032] S22. Given the interaction effects between nodes and the complexity of the paths, α is defined as a matrix:

[0033]

[0034] where C path (i,j) represents the path cost from node i to node j, L i and L j represent the load of node i and node j respectively;

[0035] Each element α in the matrix ij reflects the specific weight adjustment from node i to node j, and the element α in the matrix ij is adjusted according to the specific situations of nodes i and j; This matrix form can capture the complex associations between different nodes in the network and reflect the global changes in the logistics network;

[0036] S23. Introduce the emergency response penalty function f(x v (t), L), which is the emergency handling cost introduced when abnormal situations occur at logistics nodes. Abnormal situations include node failure, warehouse explosion, and traffic congestion; The emergency response penalty function adjusts the cost according to the load L t and the current node or path state x v (t) to reflect the additional cost in emergency situations. Its formula is as follows:

[0037] f(x v (t), L) = λ1·I(x v (t) = fail) + λ2·max(0, L t - L max )

[0038] where I(x v (t) = fail) is the indicator function, which is 1 when node v fails and 0 otherwise. λ1 represents the emergency handling cost when the node fails, and λ2 represents the cost of warehouse explosion when the maximum load L max is exceeded. L t represents the load of the node at time t, and L maxIndicates the maximum load capacity of the node or path.

[0039] Preferably, in step S3, the local path update mechanism specifically includes the following:

[0040] When a certain node fails, that is, x v (t) = 0, the chain routing algorithm only updates the affected paths instead of recalculating globally:

[0041] P new = P old - P failed + P recaluated

[0042] Among them, P new is the updated optimal path; P old represents the original, old optimal path; P failed is the failed path; P recaluated is the newly added path.

[0043] The present invention also provides a dynamic logistics network path optimization system based on the chain state routing algorithm, including:

[0044] A data acquisition module for obtaining information on the node status, path weights, and loads of the logistics network in real time;

[0045] A routing calculation module that optimizes the logistics path based on the chain routing algorithm, calculates the cost of the path, and performs dynamic updates;

[0046] A resource scheduling module that intelligently allocates logistics resources according to the optimized path and resource requirements, and executes path adjustment.

[0047] Preferably, in the routing calculation module, the cost function includes a transportation cost prediction model and an emergency response penalty mechanism, and dynamically adjusts the path selection in combination with node failures and path congestion conditions.

[0048] The present invention also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer executes the instructions, the computer realizes the dynamic logistics network path optimization method based on the chain state routing algorithm.

[0049] Compared with the prior art, the beneficial effects of the present invention: Compared with the traditional logistics model, when dealing with the path problem containing failed nodes, the present invention combines the dynamic chain routing algorithm with programmable logistics technology, has obvious advantages in path calculation cost and response time, can effectively optimize the logistics path planning, the dynamic chain routing algorithm significantly reduces the calculation and response overhead, reduces the transportation cost, and improves the robustness and resource allocation efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention.

[0051] In the accompanying drawings:

[0052] Figure 1 is a schematic flowchart of the dynamic logistics network path optimization method based on the chain state routing algorithm of the present invention;

[0053] Figure 2 is a comparison chart of the path costs between the path planning scheme finally output by the present invention and those of two other traditional logistics models;

[0054] Figure 3 is a comparison chart of the response times between the path planning scheme finally output by the present invention and those of two other traditional logistics models. Specific Embodiments

[0055] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and are not used to limit the present invention.

[0056] Embodiment: As Figure 1 shown, a dynamic logistics network path optimization method based on the chain state routing algorithm takes 64 logistics nodes in a logistics network in a certain region of China as an example, calculates its path planning scheme and the final cost, and compares it with two other traditional logistics models. The specific steps are as follows:

[0057] Step 1: Construct a logistics network topology graph, and collect the logistics network state information (node availability, edge weight, load, etc.) in real time through the chain state routing algorithm to construct a dynamic topology graph.

[0058] Step 1.1: Definition of node state

[0059] In order to determine whether node v is valid, a piecewise function is used to represent its state, and the determination is made based on conditions such as the working state (such as whether there is a fault) and load condition of the node. Based on information such as the health state and current load of the node, a piecewise function is defined such that it is 1 when the node is valid and 0 when the node is invalid.

[0060] Common conditions include:

[0061] 1. Whether the node is faulty: If node v is working properly, it is valid; if node v fails (such as hardware failure, communication interruption, etc.), it is invalid.

[0062] 2. Whether the node load exceeds the threshold: If the load L of the node t exceeds the preset maximum load Lmax If so, the node is considered overloaded and may be in an invalid state.

[0063] The validity of node v is determined by the following piecewise function:

[0064] where, when the state index x v (t) of node v at time t is greater than or equal to the threshold θ, the value of the function I(x v (t)) is 1, which means that node v is valid in this situation; when the state index x v (t) of node v at time t is less than the threshold θ, the value of the function I(x v (t)) is 0; that is, node v is invalid at this time.

[0065] To determine whether node v is valid at time t, a mathematical formula based on multiple state information is constructed, and the formula form is as follows:

[0066]

[0067] where, L v (t) represents the load of the node at time t, and L max represents the maximum load capacity of the node or path; T v (t) represents the time for the warehouse node to process goods, that is, the time required for the logistics vehicle to enter and leave the warehouse; T max represents the maximum allowed processing time.

[0068] The first exponential term is the load term, which is used to reflect the load situation of the node. When the node load L v (t) approaches the maximum load L max , this term approaches 0, meaning that the node is approaching the failure state; when the load is low, this term approaches 1, indicating that the node is normal.

[0069] The second exponential term is the event processing term, which is used to reflect the impact of the warehouse processing vehicle goods time. When the warehouse processing vehicle goods time is long, thus, when the warehouse processing time is too long, the node validity X v (t) will be significantly reduced, indicating that the node may perform poorly in the logistics network.

[0070] The processing time T v (t) of the node warehouse is related to multiple factors, such as the number of loading and unloading workers, the number of vehicles, the processing capacity of the warehouse, etc. In this embodiment, a function is used to represent the processing time and these influencing factors are incorporated:

[0071]

[0072] Among them, N w (t) represents the number of workers in the warehouse at time t, and N c (t) represents the number of vehicles being processed simultaneously at time t, and C w (t) represents the processing capacity of the warehouse, that is, the maximum number of vehicles processed per unit time; F(N c (t)) is the vehicle processing function, indicating the number of vehicles N c (t) processed by the warehouse at time t and its impact on the processing time.

[0073] As N c (t) increases, the processing time usually increases. Therefore, F(N c (t)) is an increasing function and is in the following form:

[0074] F(N c (t)) = a·N c (t) b

[0075] where a and b are constants, representing the non - linear impact of the number of vehicles processed on time; and b ≥ 1. When the number of vehicles increases, the processing time increases at a faster rate.

[0076] The function of the number of workers and the warehouse processing capacity G(N w (t), C w (t)): This function represents the impact of the number of workers and the warehouse processing capacity on the processing time. As the number of workers N w (t) increases or the warehouse processing capacity C w (t) improves, the processing time T v (t) will decrease, which is expressed as:

[0077] G(N w (t), C w (t)) = δ·N w (t)+ε·C w (t)

[0078] where δ and ε are constants, reflecting the contributions of the number of workers and the warehouse processing capacity.

[0079] Step 1.2: Calculation of transportation path weights

[0080] The transportation cost of the path changes with time and consists of two parts: static cost and dynamic cost:

[0081] w ij (t) = d ij +c ij (t)

[0082] where: d ij is the basic transportation cost (fuel consumption, toll, etc.) from node i to node j; c ij is the dynamic adjustment cost based on factors such as real-time traffic flow, freight volume, and weather.

[0083] Step 2: Dynamic path optimization. The core goal of path planning is to complete the goods delivery in the shortest time at the lowest cost, and at the same time have the ability of dynamic adjustment to ensure that the system can adapt to emergencies (such as node failure, traffic jam).

[0084] Step 2.1: Calculate the optimal solution of the path using the transportation cost prediction model

[0085] The transportation cost prediction model is used to measure the total transportation cost of the path, and its optimization goal is:

[0086]

[0087] where P is the set of paths from the source point s to the target point t; α is a parameter that weighs the normal transportation cost and the emergency handling cost; f(x v (t), L) is a penalty function used to optimize the path selection and avoid overloaded or failed nodes.

[0088] Considering the interactive influence between nodes and the complexity of the path, in this embodiment, α is defined as a matrix. For example, assume that α is a |V|×|V| matrix, representing the adjustment coefficient between every two nodes. Each element α in the matrix ij reflects the specific weight adjustment from node i to node j.

[0089]

[0090] where:

[0091]

[0092] C path (i, j) represents the path cost from node i to node j;

[0093] L i and L j respectively represent the load of node i and node j;

[0094] The element α in the matrix ij is adjusted according to the specific situations of nodes i and j. This matrix form can capture the complex associations between different nodes in the network and reflect the global changes in the logistics network.

[0095] Step 2.2: Introduction of the emergency response penalty function

[0096] The emergency response penalty function f(xv (t), L) is the emergency handling cost introduced when abnormal situations occur at logistics nodes (such as node failure, warehouse congestion, traffic jams). This function adjusts the cost according to the load L t and the current node or path status x v (t) to reflect the additional costs in emergency situations. The formula is as follows:

[0097] f(x v (t), L) = λ1·I(x v (t) = fail) + λ2·max(0, L t -L max ).

[0098] Among them, I(x v (t) = fail) is an indicator function, λ1 is the emergency handling cost when the node fails; λ2 is the warehouse congestion cost when exceeding the maximum load L max ; L t represents the load of the node at time t, and L max represents the maximum load capacity of the node or path.

[0099] Step 2.3: Path local update strategy

[0100] When a certain node fails, that is, x v (t) = 0, the chain routing algorithm only updates the affected paths instead of recalculating globally:

[0101] P new = P old - P failed + P recaluated

[0102] Among them, P new is the updated optimal path; P old represents the original, old optimal path; P failed is the failed path; P recaluated is the newly added path.

[0103] The present invention also provides a dynamic logistics network path optimization system based on the chain state routing algorithm, including:

[0104] A data acquisition module for real-time obtaining information on the node status, path weights, and loads of the logistics network;

[0105] A routing calculation module that optimizes the logistics path based on the chain routing algorithm, calculates the cost of the path, and performs dynamic updates; in the routing calculation module, the cost function includes a transportation cost prediction model and an emergency response penalty mechanism, and dynamically adjusts the path selection in combination with node failures and path congestion conditions;

[0106] A resource scheduling module, according to the optimized path and resource requirements, intelligently allocates logistics resources and performs path adjustment.

[0107] The present invention also provides a storage medium, in which instructions are stored. When a computer reads the instructions, the computer executes the programmable logistics network dynamic path optimization method based on the chained state routing algorithm.

[0108] The present invention also provides an electronic device, including a processor and the medium, and the processor executes the instructions in the storage medium.

[0109] As described in the above embodiments, the path planning scheme finally output is compared with the path costs and response times of the other two traditional logistics models, as Figure 2 and Figure 3 shown. It can be concluded that when facing the path problem with failed nodes, the dynamic chained routing algorithm has obvious advantages both in terms of path calculation cost and response time. Compared with the Dijkstra algorithm, the dynamic chained routing algorithm significantly reduces the calculation and response overhead; compared with the Bellman-Ford algorithm, although their path calculation costs are similar, the dynamic chained routing algorithm has more advantages in terms of response time. Therefore, the dynamic chained routing algorithm shows higher efficiency and better performance in a dynamic network environment with more failed nodes.

[0110] Finally, it should be noted that the above are only preferred examples of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A dynamic logistics network path optimization method based on a chain state routing algorithm, characterized in that It includes the following steps: S1. Based on the chain state routing protocol, the operating status of logistics nodes is monitored in real time through a distributed state acquisition mechanism, a node status evaluation model is constructed, and a dynamic elastic network topology structure is generated; S2. When changes occur in the logistics network nodes, a dynamic chain routing algorithm is adopted, and the logistics path is optimized by combining a transportation cost prediction model, an emergency response penalty mechanism, and a path adjustment weight coefficient; S3. A local path update mechanism is used to optimize and adjust the affected paths under abnormal node states.

2. The dynamic logistics network path optimization method based on the chained state routing algorithm according to claim 1, characterized in that: In step S1, the operating status of the logistics nodes includes parameters such as the load level and processing timeliness.

3. The dynamic logistics network path optimization method based on the chain state routing algorithm according to claim 1, wherein: In step S1, the construction of the logistics node status evaluation model includes the following steps: S11. A piecewise function is used to judge the node status, and the judgment is made based on the working status and load conditions of the node; based on the health status and current load information of the node, a piecewise function is defined such that it is 1 when the node is valid and 0 when the node is invalid. The effectiveness of node v is judged through the following piecewise function: Among them, when the state index x v (t) of node v at time t is greater than or equal to the threshold θ, the value of the function I(x v (t)) is 1. In this situation, it means that node v is valid at this time. When the state index x v (t) of node v at time t is less than the threshold θ, the value of the function I(x v (t)) is 0, that is, node v is invalid at this time; S12. By constructing a mathematical formula based on various state information, it is judged whether node v is valid at time t. The formula form is as follows: Among which L v (t) represents the load of the node at time t, and L max represents the maximum load capacity of the node or path; T v (t) represents the time for the warehousing node to process goods, that is, the time required for the logistics vehicle to enter and leave the warehouse; T max represents the maximum allowed processing time; S13. Define T v (t) is the processing time of the node repository, and its expression form is as follows: Among them, N w (t) represents the number of workers in the warehouse at time t, and N c (t) represents the number of vehicles being processed simultaneously at time t, and C w (t) represents the processing capacity of the warehouse, that is, the maximum number of vehicles processed per unit time; F(N c (t)) is the vehicle processing function, representing the impact of the number of vehicles N c (t) processed by the warehouse at time t on the processing time; S14. Define F(N c (t)) as the vehicle processing function, representing the number of vehicles N c (t) processed by the warehouse at time t and its influence on the processing time, and its expression is as follows: F(N c (t)) = a·N c (t) b where a and b are constants, representing the non-linear influence of the vehicle processing quantity on time; S15, the function G(N w (t), C w (t)) of the number of workers and the warehouse handling capacity, which represents the influence of the number of workers and the warehouse handling capacity on the processing time; as the number of workers N w (t) increases or the warehouse handling capacity C w (t) improves, the processing time T v (t) will decrease, expressed as: G(N w (t),C w (t)) = δ·N w (t) + ε·C w (t) where δ and ε are constants, reflecting the contributions of the number of workers and the warehouse processing capacity.

4. A dynamic logistics network path optimization method based on a chained state routing algorithm according to claim 1, characterized in that: In step S2, the path weight transportation cost prediction model, the emergency response penalty mechanism, and the path adjustment weight coefficient include the following steps: S21. A transportation cost prediction model C(P) is constructed to optimize the entire transportation link. The cost function form is as follows: Among them, P is the set of paths from the source point s to the target point t, α is a parameter for weighing the normal transportation cost and the emergency handling cost, and f(x v (t), L) is a penalty function used to optimize the path selection; L represents the load level in the logistics activity, and w ij (t) represents the path weight, i.e., the cost, at time t, which is expressed as: w ij f(t) = d ij + c ij f(t) where d ij represents the distance or basic transportation cost from node i to node j; c ij (t) represents the dynamic adjustment cost at time t; S22. Considering the interactive influence between nodes and the complexity of the path, α is defined as a matrix: Among them, C path (i, j) represents the path cost from node i to node j, and L i and L j represent the load of node i and node j respectively; Each element α in the matrix ij reflects the specific weight adjustment from node i to node j, and the element α in the matrix ij is adjusted according to the specific situations of nodes i and j; S23. Introduce the emergency response penalty function f(x v (t), L), which is the emergency handling cost introduced when abnormal conditions occur at the logistics node. The emergency response penalty function adjusts the cost according to the load L t and the current node or path state x v (t) to reflect the additional cost in the emergency situation. The formula is as follows: f(x v (t),L) = λ1·I(x v (t) = fail) + λ2·max(0,L t -L max ) where I(x v (t) = fail) is an indicator function, which is 1 when node v fails and 0 otherwise. λ1 represents the emergency handling cost when the node fails, and λ2 represents the cost of running out of stock when the load exceeds the maximum load L max . L t represents the load of the node at time t, and L max represents the maximum load capacity of the node or path.

5. A dynamic logistics network path optimization method based on a chained state routing algorithm according to claim 1, characterized in that: In step S3, the specific steps of the local path update mechanism are as follows: When a certain node fails, i.e., x v (t) = 0, the chain routing algorithm only updates the affected paths instead of recalculating globally: P new = P old -P failed +P recaluated Among them, P new is the updated optimal path; P old represents the original, old optimal path; P failed is the invalid path; P recaluated is the newly added path.

6. A dynamic logistics network path optimization system based on a chained state routing algorithm, characterized in that, It includes: A data acquisition module for obtaining the node status, path weight, and load information of the logistics network in real time; A routing calculation module that optimizes the logistics path based on the chain routing algorithm, calculates the cost of the path, and performs dynamic updates; A resource scheduling module that intelligently allocates logistics resources according to the optimized path and resource requirements, and executes path adjustment.

7. A dynamic logistics network path optimization system based on a chained state routing algorithm according to claim 6, characterized in that: In the routing calculation module, the cost function includes a transportation cost prediction model and an emergency response penalty mechanism, and dynamically adjusts the path selection in combination with node failure and path congestion conditions.

8. A computer-readable storage medium, characterized in that, Stored thereon are computer-executable instructions. When the computer executes the instructions, it enables the computer to implement a dynamic logistics network path optimization method based on the chain state routing algorithm.

Images