A method and system for intelligent logistics transfer route planning based on ant colony algorithm

Abstract

This application relates to the field of intelligent logistics technology, and in particular to an intelligent logistics transfer path planning method and system based on ant colony optimization (ACO). The method includes: acquiring operational status data of transfer nodes and transportation paths in a logistics network; for any transfer node in the logistics network, analyzing the stacking level of the backlog queue length and the sorting speed of the transfer node to construct the throughput lag; for any transportation path in the logistics network, constructing the path comprehensive impedance; generating a dynamic gain factor; using the product of the dynamic gain factor and the basic heuristic information in the probability transition formula of the ant colony optimization algorithm as modified heuristic information, and using the ant colony optimization algorithm to plan the logistics transfer transportation path based on the modified heuristic information. This application has the effect of improving path planning efficiency.

Description

Technical Field

[0001] This application relates to the field of intelligent logistics technology, and in particular to an intelligent logistics transfer path planning method and system based on ant colony algorithm. Background Technology

[0002] Intelligent logistics transshipment route planning refers to the algorithmic decision-making process used in complex logistics networks encompassing multiple shipping points, transit hubs, and receiving points. This process comprehensively considers multiple constraints, including transportation distance, transportation cost, transit node capacity, and delivery time. With the explosive growth of logistics volume, the topology of logistics networks is becoming increasingly complex and dynamic. Traditional scheduling methods based on human experience or static models can no longer meet the demands of processing massive orders.

[0003] In existing technologies, the Ant Colony Optimization (ACO) algorithm is typically used to solve this type of transportation path optimization problem. As a biomimetic evolutionary algorithm that simulates the foraging behavior of ants in nature, the Ant Colony Optimization (ACO) algorithm, with its unique positive feedback mechanism and distributed parallel computing capabilities, demonstrates strong robustness in solving combinatorial optimization problems such as the Traveling Salesman Problem (TSP).

[0004] However, real-world logistics and transportation scenarios often face dynamic changes such as sudden traffic congestion or backlogs at transit nodes. In such a complex, high-dimensional environment, directly applying the original ant colony algorithm can lead to significant blindness in the initial search phase. This is because the algorithm typically uses only the reciprocal of distance as static heuristic information in the early stages, and the initial pheromone concentration differences along various transport paths are minimal. This blindness often results in the algorithm performing numerous ineffective searches on inferior transport paths, leading to slow convergence, low computational efficiency, and an inability to meet the rapid response requirements of intelligent logistics systems in the face of emergencies, ultimately resulting in sudden traffic congestion or backlogs at transit nodes. Summary of the Invention

[0005] To address the issues of slow convergence speed, sudden congestion, and operational bottlenecks at transit nodes in traditional transportation route planning algorithms, this application provides an intelligent logistics transfer route planning method and system based on the ant colony algorithm.

[0006] Firstly, this application provides an intelligent logistics transfer route planning method based on ant colony algorithm, employing the following technical solution: A smart logistics transfer route planning method based on ant colony algorithm includes: acquiring the operational status data of transfer nodes and transportation routes in the logistics network, wherein the operational status data of transfer nodes includes: the length of the backlog queue and the sorting speed of the sorting equipment; and the operational status data of transportation routes includes: the traffic congestion coefficient on the transportation route. For any transit node in the logistics network, the throughput lag is constructed by analyzing the stacking level of the backlog queue length and the sorting speed of the transit node. The throughput lag is positively correlated with the stacking level and the sorting speed. For any transportation path in the logistics network, construct the path comprehensive impedance. The path comprehensive impedance is positively correlated with the physical distance of the transportation path, the real-time traffic congestion coefficient, and the throughput lag of the nodes at both ends of the transportation path. A dynamic gain factor is generated, which is negatively correlated with the overall impedance of the current transportation path; The product of the dynamic gain factor and the basic heuristic information in the probability transition formula of the ant colony algorithm is used as the modified heuristic information. Based on the modified heuristic information, the ant colony algorithm is used to plan the logistics transfer and transportation route.

[0007] By analyzing the backlog queue length and sorting speed at transit nodes, a throughput lag level is constructed, elevating the static assessment of whether a transit node is congested to a dynamic quantitative indicator of whether it is in a nonlinear instability critical state. For transit nodes, if the sorting speed is fast, then congestion at such nodes will have a greater impact on the entire logistics network.

[0008] By assessing throughput latency, the system can identify potential risks before nodes become fully congested, thus preventing logistics vehicles from being directed to impending congestion at transfer nodes and reducing hidden waiting times. Subsequently, by constructing a path-wide impedance that is positively correlated with physical distance, real-time traffic congestion coefficient, and throughput latency at both ends of the path, risk assessment of the path is achieved. This avoids the problem of selecting paths solely based on segment length or road conditions while ignoring the processing capacity of transfer nodes, making path evaluation more closely reflect real logistics efficiency and effectively eliminating false optimal paths that are short but unusable. Finally, a dynamic gain factor is generated using the path-wide impedance to correct the heuristic information in the probability transition formula of the ant colony algorithm. This enables the ant colony algorithm to perceive the real environment of the logistics network, adjusting its search direction based on dynamic changes in the logistics network, reducing the blind early search and slow convergence speed of traditional algorithms, thereby reducing the risk of transfer node congestion.

[0009] Optionally, it also includes: preprocessing the operational status data of transit nodes and transportation routes, with the preprocessing operation including at least: filling in missing values ​​in the data using Lagrange interpolation.

[0010] Lagrange interpolation is used to fill in missing values ​​in the data to ensure the continuity of the data sequence.

[0011] Optionally, the step of analyzing the stacking level of the backlog queue length of the transfer node and the sorting speed of the transfer node to construct the throughput lag degree includes: calculating the ratio of the stacking level to the preset maximum processing capacity of the transfer node, and using the logarithm of the ratio and the sorting speed of the transfer node as the current throughput lag degree of the transfer node.

[0012] Stacking level reflects the current level of cargo accumulation at a transit node. Comparing the stacking level with the designed maximum processing capacity reflects how close the current cargo accumulation at the transit node is to the maximum processing capacity, and thus reflects the risk of future congestion at the node.

[0013] Optionally, the steps for obtaining the backlog level of the relay node backlog queue length include: for any given time, constructing a time analysis window, and using the average backlog queue length corresponding to each time within the time analysis window as the backlog queue length backlog level at that time.

[0014] Calculate the average length of goods queues within a time analysis window to reduce the impact of transient and occasional abnormal data on the calculation of throughput lag and improve the robustness of the system.

[0015] Optionally, for any given moment, the average sorting speed of the transfer node at each moment within the corresponding time analysis window is taken as the sorting speed of the transfer node.

[0016] By calculating the average sorting speed within the time analysis window, a stable benchmark is provided for evaluating the actual throughput capacity of a node. This avoids misjudging the node's processing capacity due to momentary equipment failure or extremely short-term operation interruptions, ensuring the reliability and robustness of sorting speed in the throughput lag calculation.

[0017] Optionally, for any transportation route in the logistics network, the steps for constructing the route comprehensive impedance include: calculating the product of the throughput lag levels of the transfer nodes at both ends of the transportation route, using the ratio of this product to the sum of the throughput lag levels of the transfer nodes at both ends of the transportation route as a control coefficient, and using the product of the control coefficient after processing by an exponential function, the physical distance of the transportation route, and the real-time traffic congestion coefficient as the route comprehensive impedance.

[0018] An exponential function is used to amplify the throughput lag at the transfer nodes at both ends of the path and map it onto the overall path impedance. This avoids the limitation of traditional algorithms that treat road segments and nodes in isolation, ensuring that the algorithm can impose a very high penalty on paths connecting congested nodes when evaluating the path, preventing goods from passing through road segments quickly but getting stuck at transfer points.

[0019] Optionally, the step of generating the dynamic gain factor includes: for any transportation path, using the ratio of the path composite impedance of the transportation path to the average of the path composite impedances of all currently selectable transportation paths in the logistics network as a first adjustment coefficient, using the logarithm of the ratio of the maximum value of the path composite impedance of all transportation paths to the path composite impedance of the transportation path as a second adjustment coefficient, and using the product of the first adjustment coefficient and the second adjustment coefficient as the dynamic gain factor of the transportation path.

[0020] A dynamic gain factor is generated by analyzing the mean and maximum path impedances in the current logistics network, and its logarithmic form is used to further amplify the difference between high-quality and low-quality paths. This provides strong positive guidance for the ants in the early stages of the algorithm's operation.

[0021] Optionally, the steps to obtain operational data of each transit node in the logistics network include: using radio frequency identification technology and infrared sensing technology to collect the number of goods entering and leaving the warehouse and the real-time queue length.

[0022] Optionally, the length of the time analysis window is 3-8 minutes.

[0023] Secondly, this application provides an intelligent logistics transfer route planning system based on ant colony algorithm, which adopts the following technical solution: An intelligent logistics transfer route planning system based on ant colony algorithm includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the intelligent logistics transfer route planning method based on ant colony algorithm described above.

[0024] The aforementioned intelligent logistics transfer route planning system based on ant colony algorithm is generated into a computer program and stored in a memory for loading and execution by a processor. Thus, a system can be built based on the memory and processor for convenient use.

[0025] This application has the following technical effects: A throughput lag level is constructed to reflect the risk of imbalance between queuing and sorting capacity at transit nodes, and this is mapped to a comprehensive path impedance. Based on this, a dynamic gain factor negatively correlated with impedance is generated to correct the heuristic information of the ant colony algorithm, giving the algorithm real-time perception of the logistics network environment. This solves the problem of false optimal paths caused by traditional algorithms ignoring node backlogs. Simultaneously, by correcting the heuristic factor, the ant colony algorithm is guided to converge quickly, reducing the number of ineffective explorations of inferior paths, improving the real-time performance of the path planning process, and reducing the risk of congestion at transit nodes. Attached Figure Description

[0026] Figure 1 This is a flowchart of an intelligent logistics transfer route planning method based on ant colony algorithm according to an embodiment of this application.

[0027] Figure 2 This is a path-comprehensive impedance diagram of each transit node in the embodiments of this application, which is mainly used to reflect the relationship between the total impedance of the transportation path and the throughput lag of the transit nodes at both ends.

[0028] Figure 3 This is a comparison diagram of the paths planned by the ant colony algorithm in this application and the paths planned by the traditional ant colony algorithm. Detailed Implementation

[0029] This application discloses an intelligent logistics transfer route planning method based on ant colony optimization. First, it utilizes IoT technology to comprehensively perceive the logistics network status and calculates the throughput lag degree of transfer nodes, reflecting the congestion trend. Then, it combines this index with road condition information to construct a comprehensive path impedance, achieving a fusion assessment of point-to-line risks. Based on this, a dynamic gain factor is generated to correct the probability transfer rules of the ant colony optimization algorithm, thereby guiding logistics vehicles to proactively avoid potentially high-risk areas, achieving global optimization and rapid response in transportation route planning. Reference Figure 1 A smart logistics transfer route planning method based on ant colony algorithm includes steps S1-S5.

[0030] S1: Obtain operational status data of transit nodes and transportation routes in the logistics network.

[0031] For each transit node, its operational status data can include: the amount of goods entering the warehouse, the amount of goods leaving the warehouse, the length of the backlog queue, and the sorting speed of the sorting equipment. Specifically, a data perception layer for the logistics network is constructed. In terms of the selection of data acquisition hardware, for logistics transit nodes (such as distribution centers and warehousing bases), industrial-grade RFID (Radio Frequency Identification) readers with an operating frequency of 902MHz-928MHz are deployed, along with high-sensitivity infrared counting sensors, to collect the amount of goods entering the warehouse, the amount of goods leaving the warehouse, the length of the backlog queue, and the sorting speed of the sorting equipment at each transit node in real time with a sampling period of 1 second. For transport vehicles, their operational status data can include: the vehicle's real-time latitude and longitude coordinates, driving speed vector, historical trajectory information, and load status, as well as the traffic congestion coefficient on the transport route. Specifically, a dual-mode positioning terminal integrating the BeiDou-3 system and GPS (Global Positioning System) and an on-board diagnostic (OBD) system are installed, with the data collection frequency set to once per second to obtain the vehicle's real-time latitude and longitude coordinates, driving speed vector, historical trajectory information, and load status. At the same time, the system accesses the public databases of the meteorological bureau and traffic management departments through API (Application Programming Interface) to obtain real-time rainfall, visibility, and traffic congestion coefficients covering the transport route.

[0032] The data collected by sensors are collectively referred to as raw data. After acquiring the raw data, standardization preprocessing is performed to eliminate interference from environmental noise and data heterogeneity. The preprocessing process includes: for missing trajectory or state data caused by signal blind spots, a polynomial function is constructed using Lagrange interpolation, and smooth filler values ​​are fitted using known data points before and after the missing points to ensure the continuity of data time sequence; secondly, the arithmetic mean and standard deviation of the sensor data are calculated using the Laida criterion, and outliers deviating from the mean by more than three times are removed, such as instantaneous jumps in latitude and longitude caused by equipment failure or load values ​​that violate physical laws, to ensure the authenticity of the data; finally, normalization processing is performed to map data of different dimensions, such as the road congestion index (usually 1-10), to... The interval is determined, and all asynchronously acquired time-series data are uniformly resampled and aligned to the standard timestamp.

[0033] S2: For any transit node in the logistics network, analyze the stacking level of the backlog queue length and the sorting speed of the transit node to construct the throughput lag degree, where the throughput lag degree is positively correlated with the stacking level and the sorting speed.

[0034] When the total volume of goods delivered to a transit node approaches its design capacity threshold, the sorting efficiency of the transit node does not decrease linearly; however, a small increase in the queue length of goods within the transit node can lead to an exponential increase in throughput time. Therefore, this embodiment constructs a transit node throughput lag indicator capable of sensitively capturing this non-linear characteristic, reflecting the degree to which the current state of the transit node deviates from its optimal operating state. For example, when a distribution center experiences a surge of packages in a short period, although it is not yet completely overwhelmed, its processing rate begins to show unstable fluctuations; this indicator should be able to quickly capture this potential congestion trend.

[0035] Specifically, for any given moment, a time analysis window of a preset duration is constructed based on that moment. The length of the time analysis window is set between 3 and 8 minutes to reflect the current sorting speed in real time, while reducing the impact of individual noise on the sorting speed calculation. In this embodiment, the time analysis window is 5 minutes, and the time window includes the data for that moment and the five minutes preceding that moment. The real-time data sequence of the backlog queue length within this window is statistically analyzed, and the mean of this sequence is calculated to reflect the level of goods accumulation; at the same time, the ratio of the total outbound volume within this time window to the window duration is calculated to obtain the average sorting speed.

[0036] Based on the above statistics, the ratio of the stacking level to the preset maximum processing capacity of the transfer node is calculated, and the logarithm of this ratio is multiplied by the sorting speed of the transfer node as the current throughput lag of the transfer node.

[0037] The formula for calculating the throughput lag of the current relay node can be expressed as: In the formula, Indicates the first The degree of throughput lag at each logistics transit node; This indicates the sorting speed of the transfer node at the current moment, that is, the average sorting speed within the time analysis window corresponding to the current moment; This indicates the maximum designed processing capacity of the relay node; This represents the average length of the backlog queue within the time window; This represents the natural logarithm function.

[0038] The ratio of the mean to the maximum designed processing capacity reflects the instability of the relay node's operation. When the relay node is in a critical state of nonlinear congestion due to the tidal effect, the queue length usually fluctuates drastically, causing the mean to increase.

[0039] Meanwhile, the formula uses the natural logarithm function to smooth the fluctuation range, which not only preserves the characteristic that the indicator increases monotonically with the increase of fluctuation, but also prevents the numerical divergence caused by individual extreme fluctuations.

[0040] A higher average sorting speed at a transit node indicates its critical role in the logistics network, meaning that congestion at that node will have a greater impact on the entire network. Therefore, average sorting speed is introduced, assigning higher index values ​​to transit nodes with larger throughput at the same stacking level. This step provides early warning of congestion and breakdown risks at transit nodes, allowing the system to identify seemingly normal but impending congestion traps.

[0041] S3: For any transportation path in the logistics network, construct the path comprehensive impedance. The path comprehensive impedance is positively correlated with the physical distance of the transportation path, the real-time traffic congestion coefficient, and the throughput lag of the nodes at both ends of the transportation path.

[0042] The efficiency of a transportation route depends not only on road conditions but also on the throughput capacity of the transfer nodes at both ends. If only the physical attributes of the road segments are considered, the algorithm may easily choose a transportation route that is short but has congested transfer nodes at both ends, resulting in goods arriving at their destination quickly but failing to be stored in the warehouse. Therefore, this embodiment projects the risk characteristics of the transfer nodes onto the connecting road segments to construct a comprehensive path impedance.

[0043] Specifically, the product of the throughput delays at the transfer nodes at both ends of the transportation path is calculated. The ratio of this product to the sum of the throughput delays at the transfer nodes at both ends of the transportation path is used as the control coefficient. The product of the control coefficient after exponential function processing, the physical distance of the transportation path, and the real-time traffic congestion coefficient is used as the comprehensive impedance of the path.

[0044] For any connected transport path in the logistics network topology, the comprehensive impedance of the path can be calculated using the following formula: In the formula, Indicates a transit node to transit node The combined impedance of the paths between them; This indicates the physical distance of the transportation route; This represents the real-time traffic congestion coefficient of the transportation route, and its value ranges from 1 to 1. Indicates a transit node The degree of sluggishness in throughput; Indicates a transit node The degree of sluggishness in throughput; Represented by natural constant An exponential function with base 0.

[0045] Combination Figure 2 and Figure 3 When the delay index of the transfer nodes at both ends of the transportation route is high, the The value of increases, and after being amplified by the exponential function, the overall path impedance grows rapidly, leading to an increase in the overall path impedance between the two transit nodes. This simulates the impact of congestion at both ends of a route on logistics efficiency during actual logistics transportation. This index maps the congestion risk of transit nodes to the resistance of the transportation path, allowing the algorithm to treat road segments and transit nodes as a whole system when evaluating the merits of transportation paths. This effectively identifies and eliminates false optimal transportation paths that are physically short but have low actual efficiency due to transit node congestion.

[0046] S4: Generate a dynamic gain factor, which is negatively correlated with the overall impedance of the current transport path.

[0047] This step primarily empowers the ant colony algorithm with environmental awareness, addressing the problem of traditional algorithms falling into blind iteration due to a lack of guidance in the early stages of the search.

[0048] Therefore, the impedance distribution of the entire network transportation path and the dynamic gain factor are utilized.

[0049] Traverse all possible transport paths in the current network, calculate the arithmetic mean of the combined impedance of the paths as a baseline, and find the maximum value of the impedance coefficient of the transport paths in the network.

[0050] Based on this, the formula for calculating the dynamic gain factor is as follows: In the formula, Indicates the transportation route The dynamic gain factor; This represents the arithmetic mean of the combined impedance of all available transport paths in the current network, and represents the average impedance level of the available transport paths. Indicates the current transportation route The overall impedance of the path; This represents the maximum impedance coefficient of all transport paths in the current network.

[0051] Indicating the transportation route A relative advantage occurs when the overall impedance of a particular transport route is significantly lower than the average impedance level of currently available transport routes. A value greater than 1 provides a positive gain. The difference between high-quality and low-quality transport paths is further amplified using a logarithmic function; the smaller the impedance, the larger the logarithmic term. Multiplying these two terms results in a very high dynamic gain factor for the low-impedance high-quality transport path. Meanwhile, the gain factor of high-impedance transport paths decays rapidly.

[0052] This step transforms the real-time congestion distribution of the network into the search power of the algorithm. At the beginning of the algorithm's operation, high-quality transportation paths are given a higher probability of selection, thereby reducing the number of times ants try inferior transportation paths and improving the efficiency of the algorithm.

[0053] S5: The product of the dynamic gain factor and the basic heuristic information in the probability transition formula of the ant colony algorithm is used as the modified heuristic information. Based on the modified heuristic information, the ant colony algorithm is used to plan the logistics transfer and transportation route.

[0054] This step uses a dynamic gain factor to adjust the state transition rules of the standard Ant Colony Optimization (ACO) algorithm and performs a specific transportation path search.

[0055] In traditional algorithms, the probability of an ant choosing the next intermediate node is mainly determined by the pheromone concentration and the static heuristic function. In this scheme, a dynamic gain factor is directly incorporated into the probability transition formula, intervening in the ant's selection behavior by changing the weight of the heuristic information.

[0056] The improved state transition probability formula is as follows: ;in, Indicates the first Only ants in the first During the next iteration, from the transit node Transfer to transit node The probability of; Indicates the first In this iteration, the transit node and transit nodes The pheromone concentration between them; Indicates a transit node and transit nodes The basic heuristic information between them (usually the inverse of the distance); Indicates the transportation route The dynamic gain factor; The pheromone importance factor is set to 1 in this embodiment based on experience. The heuristic function importance factor is set to 3 in this embodiment based on empirical values ​​to enhance the role of heuristic information. Indicates the first The set of next-hop relay nodes that an ant is currently allowed to access, i.e., adjacent relay nodes outside the tabu list.

[0057] Combination Figure 2 and Figure 3 , As a multiplicative factor, it directly acts on the original heuristic information. Under the influence of the dynamic gain factor, even if the physical distance along a certain transportation path is relatively short, If the value is large, and the overall impedance of the transportation path is high, resulting in a small dynamic gain factor, then... The value will decrease, thus forcibly reducing the probability of ants choosing that transport path. Conversely, for transport paths that are physically longer but unobstructed, a dynamic gain factor can be used. The amplification effect increases the probability of it being selected, ultimately finding the path with the minimum overall impedance.

[0058] The specific planning process of the ant colony algorithm is as follows: At the beginning of a planning cycle, the system initializes the ant colony parameters (including the number of ants, the number of iterations, the pheromone evaporation coefficient, etc.); then, according to steps two to four, the throughput lag of each transit node in the entire network, the path impedance of each segment, and the dynamic gain factor are calculated sequentially; subsequently, virtual ants are released to search for transportation paths. The ants select the next transit node according to the improved probability transition formula until a complete delivery transportation path is constructed; after completing one iteration, the pheromone concentration on the transportation path is updated according to the total cost of the transportation path, and the above process is repeated until the maximum number of iterations or the convergence condition is met; finally, the one or more transportation paths with the highest fitness are output as the optimal scheduling scheme, and the instructions are sent to the logistics scheduling center through the wireless network to achieve precise guidance of vehicles and goods. This method, by dynamically adjusting the transition probability, enables the algorithm to quickly suppress the search probability of inferior transportation paths when there is a sudden congestion or backlog of transit nodes in the logistics network, guiding the ant colony to quickly converge to the robust transportation path with the lowest overall impedance.

[0059] This application also discloses an intelligent logistics transfer route planning system based on ant colony algorithm, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent logistics transfer route planning method based on ant colony algorithm according to this application is implemented.

[0060] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0061] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for intelligent logistics transfer route planning based on ant colony algorithm, characterized in that, Obtain operational status data for transit nodes and transportation routes in the logistics network. The operational status data for transit nodes includes: backlog queue length and sorting speed of sorting equipment; the operational status data for transportation routes includes: traffic congestion coefficient on the transportation route. For any transit node in the logistics network, the throughput lag is constructed by analyzing the stacking level of the backlog queue length and the sorting speed of the transit node. The throughput lag is positively correlated with the stacking level and the sorting speed. For any transportation path in the logistics network, construct the path comprehensive impedance. The path comprehensive impedance is positively correlated with the physical distance of the transportation path, the real-time traffic congestion coefficient, and the throughput lag of the nodes at both ends of the transportation path. A dynamic gain factor is generated, which is negatively correlated with the overall impedance of the current transportation route. The product of the dynamic gain factor and the basic heuristic information in the probability transition formula of the ant colony algorithm is used as the modified heuristic information. Based on the modified heuristic information, the ant colony algorithm is used to plan the logistics transfer transportation route.

2. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 1, characterized in that, Also includes: The operational status data of transit nodes and transportation routes are preprocessed. The preprocessing operations include at least filling in missing values ​​in the data using Lagrange interpolation.

3. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 1, characterized in that, The steps for analyzing the backlog queue length of the transfer node and the sorting speed of the transfer node to construct the throughput lag include: calculating the ratio of the backlog level to the preset maximum processing capacity of the transfer node, and using the logarithm of the ratio multiplied by the sorting speed of the transfer node as the current throughput lag of the transfer node.

4. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 3, characterized in that, The steps for obtaining the backlog level of the relay node backlog queue length include: for any given time, constructing a time analysis window, and taking the average backlog queue length corresponding to each time within the time analysis window as the backlog queue length backlog level at that time.

5. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 3, characterized in that, For any given moment, the average sorting speed of the transfer node within the corresponding time analysis window is taken as the sorting speed of the transfer node.

6. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 1, characterized in that, For any transportation route in the logistics network, the steps to construct the comprehensive impedance of the route include: calculating the product of the throughput lag of the transfer nodes at both ends of the transportation route, using the ratio of this product to the sum of the throughput lag of the transfer nodes at both ends of the transportation route as the control coefficient, and using the product of the control coefficient after processing by an exponential function, the physical distance of the transportation route, and the real-time traffic congestion coefficient as the comprehensive impedance of the route.

7. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 1, characterized in that, The steps for generating the dynamic gain factor include: for any transportation path, using the ratio of the path composite impedance of the transportation path to the average path composite impedance of all currently selectable transportation paths in the logistics network as a first adjustment coefficient, using the logarithm of the ratio of the maximum path composite impedance of all transportation paths to the path composite impedance of the transportation path as a second adjustment coefficient, and using the product of the first adjustment coefficient and the second adjustment coefficient as the dynamic gain factor of the transportation path.

8. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 1, characterized in that, The steps to obtain operational data for each transit node in the logistics network include: using radio frequency identification (RFID) and infrared sensing technologies to collect data on the number of goods entering and leaving the warehouse and the real-time queue length.

9. The intelligent logistics transfer route planning method based on ant colony algorithm according to claim 5, characterized in that, The time analysis window is 3-8 minutes long.

10. An intelligent logistics transfer route planning system based on ant colony algorithm, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement an intelligent logistics transfer path planning method based on ant colony algorithm according to any one of claims 1-9.

Images