A logistics management method and device in multimodal transport

Abstract

The present invention discloses a logistics management method and device for multimodal transport, relating to the field of logistics management technology. The method comprises: obtaining the transport starting point and transport destination of goods to be transported in the multimodal transport, and obtaining the transport nodes between the transport starting point and the transport destination, as well as the transport mode between each transport node; determining an initial transport path between the transport starting point and the transport destination according to preset rules based on the transport starting point, the transport destination, the transport nodes, and the transport mode between each transport node; establishing a target optimization function for the transport of the goods to be transported, encoding the initial transport path to obtain a corresponding initial fish swarm, and iteratively optimizing the transport path using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold, thereby obtaining the optimal transport path in the multimodal transport. The present invention solves the problem that the existing methods for determining the optimal path in multimodal transport are not ideal.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a logistics management method and device in multimodal transport. Background Art

[0002] Intermodal transport refers to the comprehensive use of two or more modes of transport, such as road transport, rail transport, and water transport, to build an efficient and convenient integrated transport system through scientific coordination, smooth conversion and close connection, so as to meet the transportation needs of goods in different scenarios.

[0003] In multimodal transport, determining the optimal route is a key component of efficient transportation. Determining the shortest route isn't simply about the shortest geographical distance; it's the optimal choice after considering multiple factors, including but not limited to the cost and time of different modes of transport. Properly planning the shortest route can reduce logistics costs, improve the timeliness of cargo transportation, and enhance the reliability of logistics services.

[0004] However, existing methods for determining the optimal path have limitations such as difficulty in handling multi-objective and high-dimensional problems, resulting in the determined transportation path being less than ideal. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a logistics management method and device in multimodal transport, aiming to solve the problem that the method for determining the optimal path in multimodal transport in the prior art is not ideal.

[0006] In one aspect, the present invention provides a logistics management method for multimodal transport, the method comprising:

[0007] Obtain the transport starting point and destination of the goods to be transported in multimodal transport, as well as the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node;

[0008] Determine the initial transportation route between the transportation starting point and the transportation end point according to the preset rules based on the transportation starting point, transportation end point, transportation nodes and the transportation mode between each transportation node;

[0009] Establish a target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, and the optimal transportation path in multimodal transport is obtained.

[0010] Furthermore, in the above-mentioned logistics management method for multimodal transport, the step of iteratively optimizing the transport path using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold comprises:

[0011] For each fish in the initial school, foraging, grouping and chasing behaviors are performed to obtain the corresponding new position;

[0012] Select the optimal target position from the corresponding new positions obtained by foraging, flocking and chasing behaviors according to their respective fitness values;

[0013] The target positions of all fish in the initial fish school are merged to form a temporary fish school. The temporary fish school is subjected to non-dominated sorting and crowding calculation. A preset number of fish are selected from the temporary fish school to form a new generation of fish school.

[0014] The new generation of fish swarms replaces the initial fish swarms and iteratively optimizes the transport path until the fitness value of the target optimization function calculated using the transport path meets the preset threshold.

[0015] Furthermore, in the above-mentioned logistics management method for multimodal transport, the step of performing foraging, grouping, and tail-chasing behaviors on each fish in the initial school of fish to obtain a corresponding new position includes:

[0016] For each fish in the initial swarm, a target position is randomly selected within the perception range, and if the target position dominates the fish's current position, the fish moves one step toward the target position;

[0017] For each fish in the initial school, find the set of partner fish in the field of view, calculate the center position of the partner fish set, and move one step towards the center position if the center position dominates the current position of the fish;

[0018] For each fish in the initial school, find the best partner fish within the field of view. If the best partner fish dominates the current position of the fish, move one step towards the best partner fish.

[0019] Furthermore, in the above-mentioned logistics management method for multimodal transport, the step of randomly selecting a target position within the sensing range for each fish in the initial fish school and moving one step toward the target position if the target position dominates the current position of the fish further comprises:

[0020] Based on the historical movement trajectory of the fish, the historical optimal direction is calculated and a random direction is generated, and the mixed direction is determined using the historical optimal direction and the random direction;

[0021] Select a target point with a set distance in the mixed direction and use the target point as the corresponding target position;

[0022] The formula for determining the mixing direction is:

[0023] ;

[0024] in, The best direction in history. is a random direction, is the weight.

[0025] Furthermore, in the above-mentioned logistics management method for multimodal transport, the step of searching for a set of partner fish within the field of view for each fish in the initial school of fish, calculating the center position of the set of partner fish, and moving one step toward the center position if the center position dominates the current position of the fish further comprises:

[0026] Find the set of partner fish in the field of view, calculate the local density of each neighbor, and calculate the comprehensive score based on the local density;

[0027] Select the target neighbor with the highest score and move one step towards the target neighbor's position if the target neighbor's position dominates the fish's current position;

[0028] The calculation formula for the comprehensive score is:

[0029] ;

[0030] in, For the j The fitness of the neighbors, is the density influencing factor, For the j The local density of neighbors.

[0031] Furthermore, in the above-mentioned logistics management method for multimodal transport, the step of searching for the best partner fish within the field of view for each fish in the initial school of fish, and moving one step toward the best partner fish if the best partner fish dominates the current position of the fish, further comprises:

[0032] Select the front of the initial fish group k The best individual is selected as the leader, and the leader fitness weight is calculated:

[0033] According to the leader's fitness weight and the leader's direction, the synthesis direction is calculated and moved towards the synthesis direction with a preset probability.

[0034] Furthermore, in the above-mentioned logistics management method for multimodal transport, the expression of the objective optimization function is:

[0035]

[0036] ;

[0037] ;

[0038] in, is the transportation cost objective function, is the transportation time objective function, i and j Represent nodes, k Indicates the mode of transport, represents the number of shipping containers, For transportation mode k The unit transportation cost, For goods at the node i and nodes j The distance between For goods at the node i and nodes j Whether the mode of transportation is used k variables, For transportation mode k Switch to transportation mode m The unit transshipment cost of the terminal, is the unit connecting transport cost, For transportation mode k Switch to transportation mode m The transit distance, For goods at the node i From the mode of transportation k Switch to m variables, is the transport speed, For transportation mode k, m Unit transit time for conversion at the terminal, is the connecting distance between the container yard and the shipper / consignee warehouse, The connecting distance between the shipper / consignee warehouse and the railway or waterway station, is the unit loading time, The unit connecting transport speed.

[0039] Another object of the present invention is to provide a logistics management device for multimodal transport, the device comprising:

[0040] An acquisition module is used to obtain the transport starting point and transport destination of the goods to be transported in multimodal transport, and obtain the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node;

[0041] A determination module is used to determine an initial transportation path between the transportation starting point and the transportation end point according to a preset rule based on the transportation starting point, the transportation end point, the transportation nodes, and the transportation mode between each transportation node;

[0042] The transportation module is used to establish the target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, thereby obtaining the optimal transportation path in multimodal transport.

[0043] Another object of the present invention is to provide a readable storage medium having a computer program stored thereon, wherein the program implements the steps of the above method when executed by a processor.

[0044] Another object of the present invention is to provide an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the steps of the above method are implemented when the processor executes the program.

[0045] The present invention obtains the transport origin and destination of the goods to be transported in multimodal transport, as well as the transport nodes between the origin and destination and the transport mode between each node. Based on the transport origin, destination, nodes, and the transport mode between each node, an initial transport path between the origin and destination is determined according to preset rules. A target optimization function for the transport of the goods to be transported is established, and the initial transport path is encoded to generate a corresponding initial fish swarm. The transport path is then iteratively optimized using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold, thereby obtaining the optimal transport path for the multimodal transport. The initial path is encoded to generate the initial fish swarm, and iterative optimization is performed. The fish swarm algorithm simulates the behavior of a fish swarm and continuously explores the transport path search space, gradually approaching the optimal solution through multiple iterations. As the iterations proceed, various aspects of the transport path (such as transport mode selection and transfer node arrangement) are continuously adjusted to ensure that the path increasingly meets the requirements of the target optimization function until the fitness value meets the preset threshold, thereby finding the relatively optimal transport path for the multimodal transport and ultimately obtaining a high-quality optimal transport path. This method solves the problem that existing methods for determining the optimal path in multimodal transport are suboptimal. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Flowchart of the logistics management method in multimodal transport in the first embodiment of the present invention;

[0047] Figure 2 4 is a structural block diagram of a logistics management device in multimodal transport according to a third embodiment of the present invention.

[0048] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0049] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate several embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0050] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be an intermediate element. When an element is referred to as being "connected to" another element, it may be directly connected to the other element or there may be an intermediate element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0052] Example 1

[0053] See also Figure 1 , which shows a logistics management method in multimodal transport in a first embodiment of the present invention, and the method includes steps S10 to S12.

[0054] Step S10: obtaining the transport starting point and transport destination of the goods to be transported in the multimodal transport, and obtaining the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node.

[0055] In complex multimodal logistics and transportation scenarios, the entire transportation process often involves the coordinated operation of multiple links and multiple modes of transportation. First, the starting point of the transportation of the goods to be transported must be determined. At the same time, the transportation destination of the goods must also be determined, that is, the destination to which the goods will ultimately be delivered. There are usually a series of transportation nodes between the transportation starting point and the transportation destination. These transportation nodes are like key stations on the transportation route. They are transfer hubs between different transportation modes, such as the handover station between rail and road transportation, and the transfer terminal between water and rail transportation. It should be noted that the transportation starting point and transportation destination are also a type of transportation node.

[0056] Furthermore, it's necessary to clearly define the transportation method between each transport node. In multimodal transport, there are many different modes of transport, including road, rail, water, and air. Different transport nodes may use different modes of transport. For example, road transport may be used from the starting point to the first transport node, rail transport from the first transport node to the second transport node, and water transport from the second transport node to the destination. Understanding the transportation method between each transport node provides a foundation for subsequent route optimization.

[0057] Step S11, according to the transportation starting point, transportation end point, transportation nodes and the transportation mode between each transportation node, an initial transportation path between the transportation starting point and the transportation end point is determined according to preset rules.

[0058] Among them, after determining the transportation starting point, transportation end point, transportation nodes and the transportation mode between each transportation node, a series of initial transportation routes that meet the actual requirements can be obtained according to the actual situation. Specifically, the available transportation nodes and the transportation mode between each transportation node can be obtained first to form their own sets, and then some nodes or transportation modes can be randomly selected from the set to form a preliminary transportation route.

[0059] Step S12, establish a target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, thereby obtaining the optimal transportation path in multimodal transport.

[0060] The objective optimization function integrates various key factors in multimodal transport into a quantifiable mathematical formula to measure the pros and cons of each transport route. In actual transportation, there are many complex and diverse factors to consider, such as transportation cost and transportation time.

[0061] Exemplarily, in an embodiment of the present invention, the expression of the objective optimization function is:

[0062]

[0063] ;

[0064] ;

[0065] in, is the transportation cost objective function, is the transportation time objective function, i and j Represent nodes respectively, k Indicates the mode of transport, represents the number of shipping containers, For transportation mode k The unit transportation cost, For goods at the node i and nodes j The distance between For goods at the node i and nodes j Whether the mode of transportation is used k variables, For transportation mode k Switch to transportation mode m The unit transshipment cost of the terminal, is the unit connecting transport cost, For transportation mode k Switch to transportation mode m The transit distance, For goods at the node i From the mode of transportation k Switch to m variables, is the transport speed, For transportation mode k, m Unit transit time for conversion at the terminal, is the connecting distance between the container yard and the shipper / consignee warehouse, The connecting distance between the shipper / consignee warehouse and the railway or waterway station, is the unit loading time, The unit connecting transport speed.

[0066] In specific implementation, the objective optimization function aims to minimize transportation costs and transportation time. In addition, in specific implementation, in order to ensure the accuracy of optimization, certain constraints can also be set. For example, the flow conservation constraint must be met, the node inflow flow and outflow flow must be consistent, only one mode of transportation can be selected between two nodes, the transportation mode of goods must be converted at most once at the node, and when the transportation mode is converted at the node, it must be consistent with the transportation mode before and after the node transshipment.

[0067] Furthermore, the initial transport paths are encoded to generate the corresponding initial fish schools. The fish school algorithm requires converting the actual transport paths into a form that can be processed by computers. Each initial transport path is encoded as a "chromosome," consisting of a series of genes. Each gene represents a key piece of information in the transport path, such as a specific transport node or the selected mode of transport. For example, a route with three transport nodes and two modes of transport might be encoded as [node A, road, node B, rail, node C]. After encoding all the initial transport paths, an initial fish school is formed. The "fish" here are actually the encoded transport paths, and the fish school is the collection of these paths. The initial fish school contains different initial path plans generated based on pre-set rules, providing a rich starting point for subsequent optimization searches.

[0068] The fish school algorithm continuously optimizes transportation routes by simulating the foraging, flocking, and chasing behaviors of fish. After each iteration, the target optimization function is used to calculate the fitness value of each "fish" (transportation route), which reflects the degree to which the route meets the optimization goal. A preset threshold is a pre-set standard. When the fitness value of a path in the fish school reaches or exceeds this threshold, it indicates that a high-quality path that meets the requirements has been found. At this point, iteration stops and this path is deemed the optimal transportation route for multimodal transport.

[0069] In summary, the multimodal transport logistics management method in the above-described embodiment of the present invention obtains the transport origin and destination of the goods to be transported in the multimodal transport, as well as the transport nodes between the origin and destination and the transport mode between each transport node. Based on the transport origin, destination, transport nodes, and the transport mode between each transport node, an initial transport path between the origin and destination is determined according to preset rules. A target optimization function for the transport of the goods to be transported is established, the initial transport path is encoded to generate a corresponding initial fish swarm, and the transport path is iteratively optimized using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold, thereby obtaining the optimal transport path for the multimodal transport. The initial path is encoded to generate the initial fish swarm, and iterative optimization is performed. The fish swarm algorithm simulates the behavior of a fish swarm and continuously explores the transport path search space, gradually approaching the optimal solution through multiple iterations. As the iterations proceed, various aspects of the transport path (such as transport mode selection and transfer node arrangement) are continuously adjusted to ensure that the path increasingly meets the requirements of the target optimization function until the fitness value meets the preset threshold, thereby finding the relatively optimal transport path for the multimodal transport and ultimately obtaining a high-quality optimal transport path. The invention solves the problem that the method for determining the optimal route in multimodal transport in the prior art is not ideal.

[0070] Example 2

[0071] This embodiment also proposes a logistics management method for multimodal transport. The logistics management method for multimodal transport in this embodiment differs from the logistics management method for multimodal transport in Example 1 in that:

[0072] The step of iteratively optimizing the transport path using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold comprises:

[0073] For each fish in the initial school, foraging, grouping and chasing behaviors are performed to obtain the corresponding new position;

[0074] Select the optimal target position from the corresponding new positions obtained by foraging, flocking and chasing behaviors according to their respective fitness values;

[0075] The target positions of all fish in the initial fish school are merged to form a temporary fish school. The temporary fish school is subjected to non-dominated sorting and crowding calculation. A preset number of fish are selected from the temporary fish school to form a new generation of fish school.

[0076] The new generation of fish swarms replaces the initial fish swarms and iteratively optimizes the transport path until the fitness value of the target optimization function calculated using the transport path meets the preset threshold.

[0077] First, for each fish in the initial school, the system simulates three real-world behaviors: foraging, schooling, and chasing, generating corresponding new positions. Foraging randomly explores the surrounding area, experimenting with factors such as transport nodes and transportation methods to generate new path variations. Schooling forces fish to graze in areas with high fitness and relatively uncrowded schools, simulating information sharing and collective intelligent selection. Chasing forces fish to track the path of the "leader" with the highest fitness within the school.

[0078] Next, the fitness value of each new position is calculated based on the objective optimization function. From the new positions obtained through foraging, swarming, and chasing, the position with the best fitness value is selected as the target position for each fish. The target positions of all fish in the initial swarm are then merged to form a temporary swarm containing numerous candidate routing solutions. Non-dominated sorting and crowding calculation are performed on this temporary swarm. Non-dominated sorting, based on multi-objective optimization, stratifies paths according to dominance relationships, screening for Pareto front solutions that are not dominated by other solutions. Crowding calculation measures the density of paths in the target space, retaining paths in sparse areas to maintain population diversity. Based on the sorting and calculation results, a preset number of high-quality fish are selected from the temporary swarm to form a new generation of fish. Finally, the new generation of fish replaces the initial swarm, and the above process of simulating behavior, selecting the optimal position, and selecting new swarms is repeated. The transportation route is iteratively optimized until the fitness value of the objective optimization function calculated for a particular transportation route reaches or exceeds a preset threshold. At this point, the resulting route is considered the optimal transportation route that meets the multimodal optimization objective.

[0079] Specifically, the steps of performing foraging, grouping, and tail-chasing behaviors on each fish in the initial school of fish to obtain a corresponding new position include:

[0080] For each fish in the initial swarm, a target position is randomly selected within the perception range, and if the target position dominates the fish's current position, the fish moves one step toward the target position;

[0081] For each fish in the initial school, find the set of partner fish in the field of view, calculate the center position of the partner fish set, and move one step towards the center position if the center position dominates the current position of the fish;

[0082] For each fish in the initial school, find the best partner fish within the field of view. If the best partner fish dominates the current position of the fish, move one step towards the best partner fish.

[0083] For foraging, each fish (or transport path) randomly selects a target location within its perception range (i.e., attempts a new path) and calculates the fitness value (e.g., total cost, total time, etc.) for that target location. If the new location's fitness is better than the current location (i.e., it "dominates" the current location, meaning in a multi-objective optimization scenario, at least one objective is superior and the others are not inferior), the fish moves one step toward the target location, generating a new path.

[0084] For schooling behavior: Each fish searches for all its companion fish (i.e., other paths) within its field of view and calculates the center position of these companion fish. If the fitness value of the center position is better than the current position, the fish moves one step toward the center.

[0085] For tailgating behavior, each fish searches for a buddy with the best fitness value within its field of view (i.e., the currently found best path). If the buddy has a better fitness value than itself, it moves one step toward it. This simulates the behavior of a school of fish following a leader, ensuring that the algorithm quickly converges to the known optimal area.

[0086] Through the synergistic effect of these three behaviors, the fish school continuously explores, gathers and follows in the iterative process, gradually approaching the global optimal solution.

[0087] Furthermore, in actual implementation, it was found that the random search of foraging behavior has low search efficiency, clustering easily leads to over-concentration, and a single leader (optimal partner fish) in tail-chasing behavior easily leads to algorithm premature maturity, which can lead to low search efficiency and accuracy. Therefore, in some preferred embodiments of the present invention, for each fish in the initial fish school, the step of randomly selecting a target position within the perception range, and if the target position dominates the current position of the fish, the step of moving one step toward the target position also includes:

[0088] Based on the historical movement trajectory of the fish, the historical optimal direction is calculated and a random direction is generated, and the mixed direction is determined using the historical optimal direction and the random direction;

[0089] Select a target point with a set distance in the mixed direction and use the target point as the corresponding target position;

[0090] The formula for determining the mixing direction is:

[0091] ;

[0092] in, The best direction in history. is a random direction, is the weight.

[0093] Specifically, the algorithm introduces a hybrid mechanism combining "historically optimal directions" and "random directions" based on the random selection of target locations during feeding behavior. First, the algorithm analyzes each fish's historical movement trajectory to extract the historically optimal direction—the direction that most significantly improved fitness during past movements. This direction reflects the successful experience accumulated by the fish during earlier explorations. Simultaneously, a completely random direction is generated to prevent the algorithm from becoming trapped in a local optimum and maintain its ability to explore the search space extensively. Subsequently, the historically optimal direction and the random direction are combined using a weighted formula. The weight determines the degree of influence of each on the hybrid direction: a larger weight indicates a stronger guiding influence of historical experience on the new direction; a smaller weight indicates a greater element of random exploration. After determining the hybrid direction, the algorithm selects a target point within a set distance along that direction as the new target location. The target location determined by the hybrid direction draws on the characteristics of previously successful paths while incorporating an element of innovative exploration. This allows the fish to efficiently utilize existing information to accelerate convergence while flexibly escaping local optima when searching for the optimal transport path, thereby enhancing the algorithm's overall optimization capabilities.

[0094] In addition, for each fish in the initial school of fish, the step of searching for a set of partner fish within the field of view, calculating the center position of the set of partner fish, and if the center position dominates the current position of the fish, then moving one step toward the center position also includes:

[0095] Find the set of partner fish in the field of view, calculate the local density of each neighbor, and calculate the comprehensive score based on the local density;

[0096] Select the target neighbor with the highest score and move one step towards the target neighbor's position if the target neighbor's position dominates the fish's current position;

[0097] The calculation formula for the comprehensive score is:

[0098] ;

[0099] in, For the j The fitness of the neighbors, is the density influencing factor, For the j The local density of neighbors.

[0100] Specifically, each fish first identifies a set of buddy fish within its field of view—that is, alternative transport path solutions. Next, for each neighbor (buddy fish), the algorithm calculates its "local density." This metric reflects the degree of crowding around the neighbor. A higher local density indicates a greater number of similar path solutions within the area, potentially indicating a localized search cluster. A lower local density indicates a less explored area, suggesting the presence of potentially high-quality solutions. An overall score is then calculated by combining the neighbor's fitness and local density. The density influence factor is an adjustable parameter that determines the degree to which local density influences the overall score. This ensures that high-scoring neighbors possess both good path quality (high fitness) and a relatively sparse search area (low local density), achieving a dual screening of high-quality, underexplored areas. Finally, the fish selects the target neighbor with the highest score. If the fitness of its position is better than the current fish's (i.e., it "dominates" the current position), the fish moves one step toward the target neighbor and updates its own transport path solution. This allows the fish to use collective wisdom to move closer to the best solution during the flocking process, and to maintain the diversity of the search through density perception, thereby finding the globally optimal transportation path more efficiently.

[0101] When the time is used, the step of searching for the best partner fish within the field of view for each fish in the initial fish school, and if the best partner fish dominates the current position of the fish, then moving one step toward the best partner fish also includes:

[0102] Select the front of the initial fish group k The best individual is selected as the leader, and the leader fitness weight is calculated:

[0103] According to the leader's fitness weight and the leader's direction, the synthesis direction is calculated and moved towards the synthesis direction with a preset probability.

[0104] Among them, the top fish with the best fitness value are selected from the initial fish group. k Individuals in the shoal are designated as "leaders," representing the best-performing transport path solutions within the current shoal. A fitness weight is then calculated for each leader. The algorithm then uses each leader's direction (i.e., the vector pointing from the current fish position to the leader's position) and its fitness weight to calculate a "composite direction" through weighted summation. This direction incorporates the guidance of multiple leaders, eliminating the need for the current fish's movement to follow a single optimal individual. Instead, it comprehensively considers the trends of multiple high-quality paths. In this way, tailgating behavior in multimodal transport path optimization can leverage the experience of multiple high-quality paths for rapid iteration while maintaining exploration flexibility, thereby enhancing the algorithm's ability to find the globally optimal transport path.

[0105] In summary, the multimodal transport logistics management method in the above-described embodiment of the present invention obtains the transport origin and destination of the goods to be transported in the multimodal transport, as well as the transport nodes between the origin and destination and the transport mode between each transport node. Based on the transport origin, destination, transport nodes, and the transport mode between each transport node, an initial transport path between the origin and destination is determined according to preset rules. A target optimization function for the transport of the goods to be transported is established, the initial transport path is encoded to generate a corresponding initial fish swarm, and the transport path is iteratively optimized using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold, thereby obtaining the optimal transport path for the multimodal transport. The initial path is encoded to generate the initial fish swarm, and iterative optimization is performed. The fish swarm algorithm simulates the behavior of a fish swarm and continuously explores the transport path search space, gradually approaching the optimal solution through multiple iterations. As the iterations proceed, various aspects of the transport path (such as transport mode selection and transfer node arrangement) are continuously adjusted to ensure that the path increasingly meets the requirements of the target optimization function until the fitness value meets the preset threshold, thereby finding the relatively optimal transport path for the multimodal transport and ultimately obtaining a high-quality optimal transport path. The invention solves the problem that the method for determining the optimal route in multimodal transport in the prior art is not ideal.

[0106] Example 3

[0107] See also Figure 2 , which shows a logistics management device for multimodal transport proposed in the third embodiment of the present invention, and includes:

[0108] The acquisition module 100 is used to obtain the transport starting point and transport destination of the goods to be transported in the multimodal transport, and obtain the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node;

[0109] The determination module 200 is used to determine the initial transportation path between the transportation starting point and the transportation end point according to the transportation starting point, the transportation end point, the transportation nodes, and the transportation mode between each transportation node according to a preset rule;

[0110] The transportation module 300 is used to establish a target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, thereby obtaining the optimal transportation path in multimodal transport.

[0111] Furthermore, in the above-mentioned logistics management device in multimodal transport, the first preset selection rule and the second preset selection rule both select character elements according to a preset selection order, and the selection order of the first preset rule is different from the selection order of the second preset rule.

[0112] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiments and will not be described in detail here.

[0113] Example 4

[0114] Another aspect of the present invention further provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the above-mentioned embodiments 1 to 2.

[0115] Example 5

[0116] On the other hand, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, the steps of the method described in any one of the above-mentioned embodiments 1 to 2 are implemented.

[0117] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device), or for use in conjunction with such instruction execution system, apparatus, or device. For purposes of this specification, a "computer-readable storage medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, apparatus, or device, or in conjunction with such instruction execution system, apparatus, or device.

[0119] More specific examples (a non-exhaustive list) of computer-readable storage media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable storage medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0120] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the aforementioned embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following technologies known in the art may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0121] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0122] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A logistics management method in multimodal transport, characterized in that: The method comprises: Obtain the transport starting point and destination of the goods to be transported in multimodal transport, as well as the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node; Determine the initial transportation route between the transportation starting point and the transportation end point according to the preset rules based on the transportation starting point, transportation end point, transportation nodes and the transportation mode between each transportation node; Establish a target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, thereby obtaining the optimal transportation path in multimodal transport; The step of iteratively optimizing the transport path using the initial fish swarm until the fitness value of the target optimization function calculated using the transport path meets a preset threshold comprises: For each fish in the initial school, foraging, grouping and chasing behaviors are performed to obtain the corresponding new position; Select the optimal target position from the corresponding new positions obtained by foraging, flocking and chasing behaviors according to their respective fitness values; The target positions of all fish in the initial fish school are merged to form a temporary fish school. The temporary fish school is subjected to non-dominated sorting and crowding calculation. A preset number of fish are selected from the temporary fish school to form a new generation of fish school. The transport path is iteratively optimized by replacing the initial fish swarm with the new generation of fish swarms until the fitness value of the target optimization function calculated using the transport path meets a preset threshold; The steps of performing foraging, grouping, and tail-chasing behaviors on each fish in the initial fish school to obtain a corresponding new position include: For each fish in the initial swarm, a target position is randomly selected within the perception range, and if the target position dominates the fish's current position, the fish moves one step toward the target position; For each fish in the initial school, find the set of partner fish in the field of view, calculate the center position of the partner fish set, and move one step towards the center position if the center position dominates the current position of the fish; For each fish in the initial school, find the best partner fish within the field of view. If the best partner fish dominates the fish's current position, move one step towards the best partner fish. The step of randomly selecting a target position within the perception range for each fish in the initial fish school, and moving one step toward the target position if the target position dominates the current position of the fish, further comprises: Based on the historical movement trajectory of the fish, the historical optimal direction is calculated and a random direction is generated, and the mixed direction is determined using the historical optimal direction and the random direction; Select a target point with a set distance in the mixed direction and use the target point as the corresponding target position; The formula for determining the mixing direction is: ； in, The best direction in history. is a random direction, is the weight.

2. The multimodal transport logistics management method according to claim 1, characterized in that: The step of searching for a set of partner fish within the field of view for each fish in the initial school of fish, calculating the center position of the set of partner fish, and moving one step toward the center position if the center position dominates the current position of the fish further comprises: Find the set of partner fish in the field of view, calculate the local density of each neighbor, and calculate the comprehensive score based on the local density; Select the target neighbor with the highest score and move one step towards the target neighbor's position if the target neighbor's position dominates the fish's current position; The calculation formula for the comprehensive score is: ； in, For the j The fitness of the neighbors, is the density influencing factor, For the j The local density of neighbors.

3. The multimodal transport logistics management method according to claim 2, characterized in that: The step of searching for the best partner fish within the field of view for each fish in the initial school of fish, and moving one step toward the best partner fish if the best partner fish dominates the current position of the fish, further comprises: Select the front of the initial fish group k The best individual is selected as the leader, and the leader fitness weight is calculated: According to the leader's fitness weight and the leader's direction, the synthesis direction is calculated and moved towards the synthesis direction with a preset probability.

4. The logistics management method in multimodal transport according to claim 1, characterized in that: The expression of the objective optimization function is: ； ； in, is the transportation cost objective function, is the transportation time objective function, i and j Represent nodes, k Indicates the mode of transport, represents the number of shipping containers, For transportation mode k The unit transportation cost, For goods at the node i and nodes j The distance between For goods at the node i and nodes j Whether the mode of transportation is used k variables, For transportation mode k Switch to transportation mode m The unit transshipment cost of the terminal, is the unit connecting transport cost, For transportation mode k Switch to transportation mode m The transit distance, For goods at the node i From the mode of transportation k Switch to m variables, is the transport speed, For transportation mode k, m Unit transit time for conversion at the terminal, is the connecting distance between the container yard and the shipper / consignee warehouse, The connecting distance between the shipper / consignee warehouse and the railway or waterway station, is the unit loading time, The unit connecting transport speed.

5. A logistics management device in multimodal transport, characterized in that: The device is used to implement the multimodal transport logistics management method according to any one of claims 1 to 4, comprising: An acquisition module is used to obtain the transport starting point and transport destination of the goods to be transported in multimodal transport, and obtain the transport nodes between the transport starting point and the transport destination and the transport mode between each transport node; A determination module is used to determine an initial transportation path between the transportation starting point and the transportation end point according to a preset rule based on the transportation starting point, the transportation end point, the transportation nodes, and the transportation mode between each transportation node; The transportation module is used to establish the target optimization function for the transportation of the goods to be transported, encode the initial transportation path to obtain the corresponding initial fish school, and use the initial fish school to iteratively optimize the transportation path until the fitness value of the target optimization function calculated using the transportation path meets the preset threshold, thereby obtaining the optimal transportation path in multimodal transport.

6. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

7. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the steps of the method according to any one of claims 1 to 4 are implemented when the processor executes the program.

Images