Steel finished product water transportation integrated stowage method and device, electronic equipment and storage medium

Abstract

According to the steel finished product water transportation integrated stowage method and device, the electronic equipment and the storage medium, the method comprises the steps that by marking key information and attributes in steel finished product order information and stowage basic configuration information, the priorities among stowage requirements can be balanced, planners can accurately follow the established priorities, and the efficiency is improved. In addition, by simplifying the making process of the logistics stowage plan, the plan making time is remarkably shortened, the transportation characteristics of sea transportation and automobile transportation are comprehensively fused in the plan making process, efficient cooperation of the sea transportation and the automobile transportation is achieved, and the logistics stowage plan making efficiency is improved. The overall optimization strategy provides a more detailed and rationalized transportation solution for enterprises, and then the transportation solution is converted into obvious economic benefits.

Description

Technical Field

[0001] The present invention relates to the field of logistics dispatching of steel enterprises, and in particular to a method, device, electronic equipment and storage medium for integrated water transportation of finished steel products. Background Art

[0002] As an economical and efficient logistics method, sea transportation is a key way for steel companies to dispatch bulk goods. Generally, the sea transportation logistics process of steel products can be divided into three main steps. First, based on the shipping notice issued by the port, planners need to select suitable steel goods that meet the demand and sequence of the ship's destination. The shipping notice lists in detail the type and quantity of goods required for each destination. Planners carefully select suitable goods for each destination based on the attributes of the steel products such as urgent shipping needs, end-customer requirements, production and delivery time. Then, considering the balance of the hull, planners will assign appropriate space for each steel product and set the loading order. In the final step, considering the storage location of steel products and the load and quantity restrictions of vehicles for different steel varieties, planners will select suitable steel products for vehicle assembly and formulate a vehicle shipping schedule.

[0003] However, traditional ocean freight steel logistics scheduling mainly relies on on-site dispatchers to manually prepare loading lists, which faces many challenges:

[0004] (1) The large amount of data and the many variables make planning time-consuming. For example, in the first step of selecting suitable steel products for loading, there is a wide range of options; in the second step, allocating the selected goods to the appropriate shipping space involves multiple choices; in the third step, the reasonable assembly of vehicles is complicated in combination with the loading sequence and vehicle loading requirements.

[0005] (2) The loading process relies too much on manual experience, making it difficult to ensure the rationality of scheduling. It is necessary to comprehensively consider multiple requirements such as urgent customers, end customers, production time, and delivery time. The priorities of these requirements are not easy to balance, making it difficult for planners to accurately follow the established priorities and ensuring the quality of the plan.

[0006] (3) Global optimization is difficult, and the coordination between sea transport and truck transport is poor. Although the three-step process simplifies the work of dispatchers, it reduces logistics efficiency and may make it difficult to adjust unreasonable plans. Especially when allocating cabins and assembling vehicles, there may be problems with cabin loading not meeting requirements or truck shipment leftovers, which can only be solved by adjusting the cargo volume. Summary of the invention

[0007] The present application provides an integrated water transport loading method, device and electronic equipment for finished steel products to solve the above-mentioned technical problems.

[0008] An embodiment of the present application provides a method for integrated water transport loading of finished steel products, the method comprising: obtaining finished steel product order information, ship information and basic loading configuration information, the ship information comprising ship batch information, the basic loading configuration information comprising information for constraining finished steel product order information; marking key information in the finished steel product order information to obtain reference order information, marking key attributes in the basic loading configuration information to obtain loading reference attributes; based on the reference order information, using a clustering algorithm model to filter the ship batch flow direction in the ship batch information to obtain a target ship batch flow direction, setting initial loading parameters for each target ship batch flow direction, and obtaining a ship batch flow direction screening result; selecting the optimal reference order information cluster from the ship batch flow direction screening result as a loadable resource pool, and loading constraints are imposed on the loadable resource pool based on the loading reference attributes, so as to maximize the loading rate of each ship flow direction and minimize the number of vehicles used as optimization goals, and establish a mixed integer programming model; solving the mixed integer programming model through a large-scale neighborhood search algorithm to generate ship loading details and vehicle loading details, and completing the integrated water transport loading of finished steel products.

[0009] In one embodiment of the present application, a clustering algorithm model is used to screen the ship batch flow direction in the ship batch information to obtain a target ship batch flow direction, including: converting the time attribute in the reference order information into the number of days relative to the operation time of the clustering algorithm model, recorded as the operation time; encoding the category attribute in the reference order information to obtain the encoding result, recorded as a numerical variable; according to the preset clustering requirements, hierarchical clustering is performed on the operation time and the numerical variable to screen the ship batch flow direction in the ship batch information to obtain the target ship batch flow direction.

[0010] In one embodiment of the present application, hierarchical clustering is performed on the operation time and the numerical variables, including: constructing an initial hierarchical structure through the operation time and the numerical variables according to a preset distance or similarity metric, forming each order into an initial cluster separately to complete initialization; calculating the similarity between all cluster pairs, selecting the two cluster pairs with the highest similarity to merge, and after each merge, updating the hierarchical structure and the similarity between clusters until a preset number of clusters is reached or the similarity between cluster pairs is lower than a preset similarity threshold.

[0011] In one embodiment of the present application, loading constraints are imposed on the loadable resource pool based on the loading reference attributes, including: determining basic loading constraints for steel product orders and goods to be loaded, the basic loading constraints including that one steel product order can be loaded on at most one vehicle and one steel product order can be loaded on at most one cabin; on the basis of the basic loading constraints, establishing vehicle loading constraints and cabin loading constraints, and ensuring the consistency of goods on the ship and on the vehicle, the vehicle loading constraints including the same-vehicle loading principle, vehicle load constraints and vehicle loading quantity constraints, the cabin loading constraints including the same-cabin loading principle, cargo area constraints and cabin load constraints.

[0012] In one embodiment of the present application, a mixed integer programming model is established with the optimization objectives of maximizing the loading rate of each flow direction of the ship and minimizing the number of vehicles used, including:

[0013]

[0014] Among them, Z is the constraint objective function; D is the destination arrival order set; I is the steel product order set in the available resource pool; T is the set of loading vehicles; W i is the weight attribute of steel product order i; W d is the expected loading weight at the destination; H id is the destination attribute of steel product order i, H id ∈{0,1}; a ij is a 0-1 variable for steel product i loaded in hold j; e t Loading steel products into vehicles; t =0,if∑ i∈I b it =0, b it is a 0-1 variable representing the steel product i loaded on vehicle t, and α and β are weight parameters.

[0015] In one embodiment of the present application, the designed mixed integer programming model is solved by a large-scale neighborhood search algorithm to generate ship allocation details and vehicle loading details, including: setting the search depth, the number of iterations and the neighborhood size of the solution, determining the basic framework of the search process, and obtaining algorithm parameters and constraints; according to the algorithm parameters and constraints, setting a first group of variables for the cargo hold loading condition of steel products to obtain a cargo hold loading initial solution, and setting a second group of variables for the cargo hold loading condition of steel products to obtain a vehicle loading initial solution; encoding and modifying the cargo hold loading initial solution to obtain a cargo hold loading neighborhood, encoding and modifying the vehicle loading initial solution to obtain a vehicle loading neighborhood; performing deletion and insertion operator processing on the cargo hold loading initial solution based on the cargo hold loading neighborhood, and performing deletion and insertion operator processing on the vehicle loading initial solution based on the vehicle loading neighborhood to obtain ship allocation details and vehicle loading details.

[0016] In one embodiment of the present application, the cabin loading initial solution is processed by deleting operators and inserting operators based on the cabin loading neighborhood, and the vehicle loading initial solution is processed by deleting operators and inserting operators based on the vehicle loading neighborhood to obtain ship allocation details and vehicle loading details, including: randomly deleting some cabin loading variables in the cabin loading initial solution to obtain the remaining cabin loading initial solution; inserting cabin loading operators into the vacant cabin loading initial solution to obtain the target cabin loading initial solution; and randomly deleting some vehicle loading variables in the vehicle loading initial solution to obtain the vacant vehicle loading initial solution, inserting A vehicle loading operator is used to obtain a target vehicle loading initial solution; the target cabin loading initial solution and the target vehicle loading initial solution are integrated into a target loading initial solution; a local search is performed on the target loading initial solution, and the target loading initial solution is refined using a specific local optimization strategy, and a locally improved target loading initial solution is iteratively output, which is recorded as an iterative loading solution; the cabin loading efficiency of the iterative loading solution is evaluated by an objective function, and a current loading solution is output; during the iteration process, the objective function values ​​of the iterative loading solution and the current loading solution are compared until an optimal loading solution is output; the optimal loading solution is decoded to obtain ship allocation details and vehicle loading details.

[0017] The embodiment of the present application provides an integrated water transport loading device for steel finished products, the device comprising: an acquisition module for acquiring steel finished product order information, ship information and loading basic configuration information, the ship information including ship batch information, the loading basic configuration information including information for constraining the steel finished product order; a marking module for marking key information in the steel finished product order information to obtain reference order information, marking key attributes in the loading basic configuration information to obtain loading reference attributes; a screening module for screening the ship batch flow direction in the ship batch information based on the reference order information using a clustering algorithm model , obtain the target ship batch flow direction, set the initial loading parameters for each target ship batch flow direction, and obtain the ship batch flow direction screening result; the loading constraint module is used to select the optimal reference order information cluster as the loadable resource pool in the ship batch flow direction screening result, and load the loadable resource pool based on the loading reference attribute to maximize the loading rate of each ship flow direction and minimize the number of vehicles used as the optimization goal, and establish a mixed integer programming model; the solution module is used to solve the mixed integer programming model through a large-scale neighborhood search algorithm, generate ship allocation details and vehicle loading details, and complete the integrated water transportation loading of finished steel products.

[0018] An embodiment of the present application provides an electronic device, including a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the above-mentioned integrated water transportation loading method for finished steel products.

[0019] Beneficial effects of the present application: The present application can balance the priorities between loading requirements by marking the key information and attributes in the finished steel product order information and the basic loading configuration information, so that planners can accurately follow the established priorities and ensure the quality of the plan. By streamlining the process of formulating the logistics loading plan, the planning time is significantly shortened. In the process of formulating the plan, the transportation characteristics of sea transportation and road transportation are fully integrated to achieve efficient coordination between the two. This overall optimization strategy provides enterprises with more detailed and rational transportation solutions, which in turn translates into obvious economic benefits.

[0020] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0022] Figure 1 It is a flow chart of an integrated water transportation loading method for finished steel products shown in an exemplary embodiment of the present application;

[0023] Figure 2 is a flowchart of a large-scale neighborhood search algorithm implementation shown in an exemplary embodiment of the present application;

[0024] Figure 3 It is a block diagram of an integrated loading device for water transportation of finished steel products shown in an exemplary embodiment of the present application;

[0025] Figure 4 It is a structural schematic diagram of an electronic device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0027] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0028] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0029] The embodiments of the present application respectively propose a method for integrated water transportation loading of finished steel products, a device for integrated water transportation loading of finished steel products, an electronic device and a storage medium, and these embodiments will be described in detail below.

[0030] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for integrated water transportation of finished steel products according to an embodiment of the present application. Figure 1 As shown, the method at least includes steps S110 to S150, which are described in detail as follows:

[0031] Step S110, obtaining finished steel product order information, ship information and basic stowage configuration information, wherein the ship information includes ship lot information, and the basic stowage configuration information includes information for constraining the finished steel product order information.

[0032] As an example, the finished steel product order information is the contract information of each finished steel product after production, including the order number, contract number, material number, product name, variety, number of pieces, weight, storage area, recipient, destination, planned delivery date, production offline time, whether it is an urgent customer, and whether it is an end customer.

[0033] As an example, the steel finished product order information includes the available quantity details and material details, as shown in Table 1, Table 1 is the available quantity details table, including order number, order number, contract number, material number, consignee, product name, brand, specification, number of pieces to be shipped, weight to be shipped, number of pieces that can be shipped, weight that can be shipped, product code, destination, unloading address, customer type, urgent customer; as shown in Table 2, Table 2 is the material details table, including customized material number, contract number, order number, order item number, order number (including item number), product code, product name Chinese, brand (steel grade), standard, material thickness, material width, material length, inner diameter, outer diameter, product grade, material form mark, warehouse area code, warehouse location number, old warehouse location number, span number, layer number, row number, column number, measurement method, material net weight, material gross weight, material theoretical weight, material actual weight, material differential weight, material count, material status;

[0034] Table 1. Details of available quantities

[0035]

[0036]

[0037] Table 2. Material list

[0038]

[0039]

[0040] In one embodiment of the present application, the ship information includes ship batch information and cabin allocation information, and the basic loading configuration information includes the upper and lower limits of the loading weight of vehicles for specific types and the upper and lower limits of the loading quantity of vehicles for specific types.

[0041] As an example, Table 3 is a ship batch information table, and the ship batch information includes ship name, ship batch number, port destination, port destination number, and planned weight. When a ship has multiple port destinations, multiple data are created, and the ship batch information is not specific to a specific space, as shown in Table 3 below:

[0042] Table 3. Ship batch information table

[0043]

[0044]

[0045] As an example, Table 4 is a cabin allocation information table. The cabin allocation information is allocated by the planner to each cabin according to the order and weight of the port destination, including the cabin number, the port destination, and the planned weight limit, as shown in Table 4 below:

[0046] Table 4. Cabin allocation information table

[0047]

[0048] As an example, Table 5 is a single vehicle load configuration table. The basic loading configuration information includes the upper and lower limits of the vehicle's loading weight for a specific type and the upper and lower limits of the vehicle's loading quantity for a specific type, as shown in Table 5 below:

[0049] Table 5. Single vehicle load configuration table

[0050]

[0051] Step S120, marking key information in the steel finished product order information to obtain reference order information, marking key attributes in the loading basic configuration information to obtain loading reference attributes.

[0052] In one embodiment of the present application, key information is marked in the finished steel product order information to obtain reference order information, which specifically includes urgent customers, end customers, production time, and delivery time.

[0053] Step S130, based on the reference order information, a clustering algorithm model is used to filter the ship batch flow directions in the ship batch information to obtain target ship batch flow directions, and initial loading parameters are set for each target ship batch flow direction to obtain a ship batch flow direction filtering result.

[0054] In one embodiment of the present application, a clustering algorithm model is used to screen the ship batch flow direction in the ship batch information to obtain a target ship batch flow direction, including: converting the time attribute in the reference order information into the number of days relative to the operation time of the clustering algorithm model, recorded as the operation time; encoding the category attribute in the reference order information to obtain the encoding result, recorded as a numerical variable; hierarchically clustering the operation time and the numerical variable according to the preset clustering requirements to screen the ship batch flow direction in the ship batch information to obtain the target ship batch flow direction.

[0055] In one embodiment of the present application, hierarchical clustering is performed on operation time and numerical variables, including: constructing an initial hierarchical structure through operation time and numerical variables according to a preset distance or similarity metric, forming each order into an initial cluster separately to complete initialization; calculating the similarity between all cluster pairs, selecting the two cluster pairs with the highest similarity to merge, and after each merge, updating the hierarchical structure and the similarity between clusters until a preset number of clusters is reached or the similarity between cluster pairs is lower than a preset similarity threshold.

[0056] As an example, the clustering operation steps of the clustering algorithm include calculating the similarity between all possible cluster pairs, selecting the two clusters with the highest similarity to merge, updating the hierarchy and the similarity between clusters after each merge, and repeating this process until a preset number of clusters is reached or the similarity between clusters is lower than a certain threshold.

[0057] As an example, the implementation steps of the clustering algorithm include data preprocessing, initialization, and clustering operations. Optionally, in this embodiment, urgent customers, terminal customers, production time, and delivery time belong to common boundary fuzzy occasions in the real world. The clustering algorithm is a fuzzy clustering algorithm including but not limited to a hierarchical fuzzy clustering algorithm, a Gaussian mixture model, and a weighted fuzzy C-means algorithm. In this embodiment, a hierarchical fuzzy clustering algorithm is selected.

[0058] As an example, the data preprocessing step of the clustering algorithm includes first preprocessing the key attributes in the clustering process, including date and time attributes (e.g., production time, delivery time) and category attributes (e.g., urgent customers, end customers). Specifically, the date and time attributes are converted into days relative to the current model operation time, and the category attributes are encoded as 0 / 1 variables; then, all data are standardized using a normalization method to facilitate subsequent clustering analysis. Optionally, the method of normalizing data adopted in this embodiment includes but is not limited to Max-Min normalization, Log normalization, etc.

[0059] As an example, the initialization step of the clustering algorithm includes selecting a suitable distance or similarity metric and constructing an initial hierarchical structure; in the initial stage, each order forms an initial cluster separately. Optionally, in this embodiment, a suitable distance or similarity metric is selected, including but not limited to Euclidean distance, Manhattan distance, and Chebyshev distance.

[0060] As an example, the clustering operation steps of the clustering algorithm include calculating the similarity between all possible cluster pairs, selecting the two clusters with the highest similarity to merge, updating the hierarchy and the similarity between clusters after each merge, and repeating this process until a preset number of clusters is reached or the similarity between clusters is lower than a certain threshold.

[0061] Step S140, select the optimal reference order information cluster as the loadable resource pool in the ship batch flow screening results, and set loading constraints on the loadable resource pool based on the loading reference attributes, so as to maximize the loading rate of each flow direction of the ship and minimize the number of vehicles used as the optimization goals, and establish a mixed integer programming model.

[0062] In one embodiment of the present application, the optimal reference order information cluster is selected from the ship batch flow screening results as the loadable resource pool. Specifically, the average value of the sum of the four attribute values ​​in the clusters of urgent customers, terminal customers, production time, and delivery time is calculated to determine the cluster with the best comprehensive performance in terms of urgent customers, terminal customers, production time, and delivery time, and this cluster is used as the loadable resource pool.

[0063] In one embodiment of the present application, loading constraints are imposed on a loadable resource pool based on loading reference attributes, including: determining basic loading constraints for steel product orders and goods to be loaded, the basic loading constraints including that one steel product order can be loaded on at most one vehicle and one steel product order can be loaded on at most one hold; on the basis of the basic loading constraints, establishing vehicle loading constraints and cabin loading constraints, and ensuring the consistency of goods on the ship and on the vehicle, the vehicle loading constraints including the same-vehicle loading principle, vehicle load constraints and vehicle loading quantity constraints, the cabin loading constraints including the same-hold loading principle, cargo area constraints and cabin load constraints.

[0064] As an example, load constraints are imposed on the loadable resource pool based on the load reference attribute, including:

[0065] The same vehicle only carries the same cargo hold, that is, the same destination cargo;

[0066]

[0067] The same vehicle can only carry goods from the same cargo area;

[0068]

[0069] Only the same type of goods are loaded on the same vehicle;

[0070]

[0071] One order of steel products can be loaded on one vehicle at most;

[0072]

[0073] One order of steel products can be loaded in one hold at most;

[0074]

[0075] If steel products are loaded on a ship, they must also be loaded on a vehicle;

[0076]

[0077] The load of each cabin shall not exceed the upper load limit;

[0078]

[0079] The load of each cabin shall not be less than the lower load limit;

[0080]

[0081] The vehicle load does not exceed the upper limit;

[0082]

[0083] The vehicle load is not less than the lower limit;

[0084]

[0085] The vehicle load quantity does not exceed the upper limit;

[0086]

[0087] The vehicle loading quantity is not less than the lower limit;

[0088]

[0089] Among them, D is the destination arrival sequence set, I is the steel product order set in the available resource pool, T is the set of loading vehicles, C is the set of steel product varieties, S is the set of steel product cargo areas, and W is the set of steel product cargo areas. i is the weight attribute of steel product order i, N i is the quantity attribute of steel product order i, K ic K is the type attribute of steel product order i ic ∈{0,1},H id is the destination attribute H of steel product order i id ∈{0,1},J is is the storage area attribute J of steel product order i is ∈{0,1},D j is the arrival destination d of cabin j, SWU j is the upper limit of the loading constraint of cabin j, SWL j is the lower limit of the loading constraint of cabin j, TWU c The upper limit of the weight of steel type c loaded on vehicles, TWL c The minimum weight limit for loading steel type c on a vehicle, TNU c The upper limit of the quantity of steel type c loaded on the vehicle, TNL c W is the lower limit of the quantity of steel type c loaded on the vehicle, d is the expected loading weight at the destination, H id is the destination attribute H of steel product order i id ∈{0,1},a ij is a 0-1 variable for steel product i loaded in hold j, bit is a 0-1 variable for the steel product i loaded on vehicle t, e t Loading steel products into vehicle t, e t =0,if∑ i∈I b it =0,f ct Type of steel products loaded on vehicle t, f ct =1, if the vehicle is loaded with steel type c.

[0090] In one embodiment of the present application, a mixed integer programming model is established with the optimization objectives of maximizing the loading rate of each flow direction of the ship and minimizing the number of vehicles used, including:

[0091]

[0092] Among them, Z is the constraint objective function; D is the destination arrival order set; I is the steel product order set in the available resource pool; T is the set of loading vehicles; W i is the weight attribute of steel product order i; W d is the expected loading weight at the destination; H id is the destination attribute of steel product order i, H id ∈{0,1}; a ij is a 0-1 variable for steel product i loaded in hold j; e t Loading steel products into vehicles; t =0,if∑ i∈I b it =0, b it is a 0-1 variable for steel product i loaded on vehicle t, and α and β are weight parameters. Specifically, α can be set to 0.4 and β can be set to 0.6.

[0093] Step S150, solving the mixed integer programming model through a large-scale neighborhood search algorithm, generating ship allocation details and vehicle loading details, and completing the integrated water transportation loading of finished steel products.

[0094] In one embodiment of the present application, a designed mixed integer programming model is solved by a large-scale neighborhood search algorithm to generate ship allocation details and vehicle loading details, including: setting the search depth, the number of iterations and the neighborhood size of the solution, determining the basic framework of the search process, and obtaining algorithm parameters and constraints; according to the algorithm parameters and constraints, setting a first group of variables for the cargo hold loading of steel products to obtain an initial cargo hold loading solution, and setting a second group of variables for the cargo hold loading of steel products to obtain an initial vehicle loading solution; encoding and modifying the initial cargo hold loading solution to obtain a cargo hold loading neighborhood, and encoding and modifying the initial vehicle loading solution to obtain a vehicle loading neighborhood; performing deletion and insertion operator processing on the initial cargo hold loading solution based on the cargo hold loading neighborhood, and performing deletion and insertion operator processing on the initial vehicle loading solution based on the vehicle loading neighborhood to obtain ship allocation details and vehicle loading details.

[0095] In one embodiment of the present application, a cabin loading initial solution is processed by deleting an operator and inserting an operator based on a cabin loading neighborhood, and a vehicle loading initial solution is processed by deleting an operator and inserting an operator based on a vehicle loading neighborhood to obtain ship allocation details and vehicle loading details, including: randomly deleting some cabin loading variables of the cabin loading initial solution in the cabin loading neighborhood to obtain the remaining cabin loading initial solution; inserting a cabin loading operator into the remaining cabin loading initial solution to obtain the target cabin loading initial solution; and randomly deleting some vehicle loading variables of the vehicle loading initial solution in the vehicle loading neighborhood to obtain the remaining vehicle loading initial solution; inserting a cabin loading operator into the remaining cabin loading initial solution to obtain the target cabin loading initial solution. The target initial solution for cabin loading is inserted into the vehicle loading operator to obtain the target initial solution for vehicle loading; the target initial solution for cabin loading and the target initial solution for vehicle loading are integrated into the target initial solution for loading; a local search is performed on the target initial solution for loading, and the target initial solution for loading is refined using a specific local optimization strategy, and the locally improved target initial solution for loading is iteratively output, which is recorded as the iterative loading solution; the cabin loading efficiency of the iterative loading solution is evaluated by the objective function, and the current loading solution is output; in the iterative process, the objective function values ​​of the iterative loading solution and the current loading solution are compared until the optimal loading solution is output; the optimal loading solution is decoded to obtain the ship allocation details and the vehicle loading details.

[0096] As an example, the specific steps to implement the large-scale neighborhood search algorithm are as follows:

[0097] Parameter initialization step: Set key algorithm parameters including but not limited to search depth, number of iterations, solution neighborhood size, etc. to establish the basic framework of the search process.

[0098] Generate initial solution step: Based on the preset parameters and constraints, construct a symbol coding system, which includes two types of 0-1 variables, representing the steel products loaded in the ship's hold and loaded in the vehicle. This step ensures that the position of each steel product in the loading plan is clearly mathematically represented. Specifically:

[0099] a. For the cargo hold loading, define a set of 0-1 variables, where each variable corresponds to the pairing of a specific steel product with a specific cargo hold. If the steel product is assigned to the cargo hold, the corresponding variable value is 1, otherwise it is 0.

[0100] b. For vehicle loading, a set of 0-1 variables is also defined, each variable corresponds to the pairing of a specific steel product with a specific vehicle. If the steel product is assigned to a vehicle, the variable value is 1, if not, it is 0.

[0101] Neighborhood definition step: The neighborhood solution set is formed by modifying the encoding of the initial solution. For steel products loaded in the hold and on the vehicle, the following definition is used to form the neighborhood:

[0102] a. Ship loading neighborhood: For each steel product loading variable in the ship, we consider moving the product from the current ship to another ship, or changing its loading status. This is achieved by changing the corresponding 0-1 encoding, that is, changing a variable from 0 to 1 to indicate loading, or from 1 to 0 to cancel loading.

[0103] b. Vehicle loading neighborhood: For the steel product loading variables on the vehicle, the possible transfer between products, between products and vehicles, or changing loading status are also considered. This is also achieved by adjusting the 0-1 variables corresponding to the steel products, modifying the variable values ​​to represent the different distribution states of the products on the vehicle.

[0104] Deletion operator and insertion operator application steps: By selecting or randomly determining specific variables for deletion operations, some parts of the initial solution are removed to explore new potential solutions. After deletion, the insertion operator is used to add the previously removed parts or new elements back to the solution, which requires evaluating their contribution to the objective function to determine the best insertion position.

[0105] a. Cabin loading: Randomly determine and delete specific cabin loading variables, set the 0-1 value to 0, indicating that the original allocation is cancelled. The insertion operator will consider reallocating or introducing new loading items into the cabin.

[0106] b. Vehicle loading: Randomly determine specific vehicle loading variables for deletion, and then explore optimized or new loading solutions through insertion operations.

[0107] The application steps of the local search algorithm are as follows: perform local search on the solutions remaining after the deletion operator is processed, refine the solution through a specific local optimization strategy, and seek local improvements of the solution.

[0108] Objective function evaluation step: The objective function is used to evaluate the performance of the solution obtained in each iteration, which includes a comprehensive evaluation of the loading rate of each flow direction of the ship and the number of vehicles used.

[0109] Solution selection and update steps: Compare the objective function values ​​of the iterative solution and the current optimal solution, and decide whether to use the iterative solution as the new current optimal solution based on the comparison result.

[0110] Termination condition determination step: Continue iterating until a pre-set termination condition is met, such as the quality of the solution no longer significantly improves, or the maximum number of iterations has been reached.

[0111] Output the optimal solution step: The optimal solution is finally selected. After decoding, the optimal solution obtains the ship allocation details and the loading details as the recommended solution for the integrated water transportation of finished steel products.

[0112] As an example, Table 6 is a shipping list, and Table 7 is a truck loading list. The shipping details calculated by the large-scale neighborhood search algorithm include ship name, ship batch number, ticket number, ticket number generation time, contract number, product name, material, specification, number of bundles, weight, outbound warehouse, port of destination, cabin number and storage location; the truck loading list includes: loading task number, car batch number generation time, contract number, order number, order item number, delivery note number, product name, specification, number of pieces, weight, outbound warehouse, storage location, destination, cabin number, ship batch number and ticket number; the shipping details and truck loading details are displayed through the system to the road transport dispatching personnel responsible for the outbound delivery of finished steel products and the sea transport dispatching personnel responsible for ship loading.

[0113] Table 6. Shipping Details

[0114]

[0115]

[0116] Table 7. Loading Details

[0117]

[0118] It should be noted that the fields "tank number" and "storage location" in Table 6 contain the key information of the storage location of the finished steel products on the ship calculated by the model. The same finished steel products "loading task number" in Table 7 are loaded on one vehicle.

[0119] exist Figure 1In the technical solution shown, by marking the key information and attributes in the finished steel product order information and the basic loading configuration information, the priorities between the loading requirements can be balanced, so that planners can accurately follow the established priorities and ensure the quality of the plan. In addition, by streamlining the process of formulating the logistics loading plan, the planning time is significantly shortened. In the process of formulating the plan, the transportation characteristics of sea transportation and road transportation are fully integrated to achieve efficient coordination between the two. This overall optimization strategy provides companies with more detailed and rational transportation solutions, which in turn translates into obvious economic benefits.

[0120] Figure 2 A flowchart for implementing a large-scale neighborhood search algorithm according to an exemplary embodiment of the present application is shown. First, parameter initialization is performed to obtain algorithm parameters and constraints. An initial solution is generated based on the algorithm parameters and constraints, and the encoding of the initial solution is modified to form a neighborhood solution set to define the neighborhood. Some variables of the initial solution in the neighborhood set are randomly deleted. After deletion, the previously removed parts or new elements are re-added to the solution using an insertion operator. A local search is performed on the remaining solutions after the deletion operator processing. The objective function is used to evaluate the performance of the solutions obtained in each iteration, and the objective function values ​​of the iterative solution and the current optimal solution are compared. Based on the comparison result, it is decided whether to use the iterative solution as the new current optimal solution. The iterative process is continued until a pre-set termination condition is met, such as when the quality of the solution is no longer significantly improved or the maximum number of iterations has been reached, and the optimal solution is finally selected.

[0121] Figure 3 FIG. 1 is a block diagram of a steel product water transport integrated loading device according to an exemplary embodiment of the present application. Figure 3As shown, according to an embodiment of the present application, a steel product water transport integrated loading device 300 includes: an acquisition module 310, a marking module 320, a screening module 230, a loading constraint module 340 and a solution module 350. The acquisition module is used to obtain the finished steel product order information, ship information and basic loading configuration information, the ship information includes ship batch information, and the basic loading configuration information includes information for constraining the finished steel product order; the marking module is used to mark key information in the finished steel product order information to obtain reference order information, and mark key attributes in the basic loading configuration information to obtain loading reference attributes; the screening module is used to screen the ship batch flow direction in the ship batch information based on the reference order information using a clustering algorithm model to obtain the target ship batch flow direction, set initial loading parameters for each target ship batch flow direction, and obtain the ship batch flow direction screening result; the loading constraint module is used to select the optimal reference order information cluster as the loadable resource pool in the ship batch flow screening result, and load the loadable resource pool based on the loading reference attribute, so as to maximize the loading rate of each ship flow direction and minimize the number of vehicles used as the optimization goal, and establish a mixed integer programming model; the solution module is used to solve the mixed integer programming model through a large-scale neighborhood search algorithm, generate ship allocation details and vehicle loading details, and complete the integrated water transportation loading of finished steel products. By marking the key information and attributes in the finished steel product order information and the basic loading configuration information, the priorities between loading requirements can be balanced, allowing planners to accurately follow the established priorities and ensure the quality of the plan. By streamlining the process of formulating logistics loading plans, the planning time is significantly shortened. In the process of formulating plans, the transportation characteristics of sea and road transportation are fully integrated to achieve efficient coordination between the two. This overall optimization strategy provides companies with more detailed and rational transportation solutions, which in turn translates into obvious economic benefits.

[0122] In one embodiment of the present application, the screening module is used to convert the time attribute in the reference order information into the number of days relative to the operation time of the clustering algorithm model, which is recorded as the operation time; the category attribute in the reference order information is encoded to obtain the encoding result, which is recorded as a numerical variable; according to the preset clustering requirements, the operation time and the numerical variable are hierarchically clustered to screen the ship batch flow direction in the ship batch information to obtain the target ship batch flow direction.

[0123] As an example, the clustering operation steps of the clustering algorithm include calculating the similarity between all possible cluster pairs, selecting the two clusters with the highest similarity to merge, updating the hierarchy and the similarity between clusters after each merge, and repeating this process until a preset number of clusters is reached or the similarity between clusters is lower than a certain threshold.

[0124] As an example, the implementation steps of the clustering algorithm include data preprocessing, initialization, and clustering operations. Optionally, in this embodiment, urgent customers, terminal customers, production time, and delivery time belong to common boundary fuzzy occasions in the real world. The clustering algorithm is a fuzzy clustering algorithm including but not limited to a hierarchical fuzzy clustering algorithm, a Gaussian mixture model, and a weighted fuzzy C-means algorithm. In this embodiment, a hierarchical fuzzy clustering algorithm is selected.

[0125] As an example, the data preprocessing step of the clustering algorithm includes first preprocessing the key attributes in the clustering process, including date and time attributes (e.g., production time, delivery time) and category attributes (e.g., urgent customers, end customers). Specifically, the date and time attributes are converted into days relative to the current model operation time, and the category attributes are encoded as 0 / 1 variables; then, all data are standardized using a normalization method to facilitate subsequent clustering analysis. Optionally, the method of normalizing data adopted in this embodiment includes but is not limited to Max-Min normalization, Log normalization, etc.

[0126] As an example, the initialization step of the clustering algorithm includes selecting a suitable distance or similarity metric and constructing an initial hierarchical structure; in the initial stage, each order forms an initial cluster separately. Optionally, in this embodiment, a suitable distance or similarity metric is selected, including but not limited to Euclidean distance, Manhattan distance, and Chebyshev distance.

[0127] As an example, the clustering operation steps of the clustering algorithm include calculating the similarity between all possible cluster pairs, selecting the two clusters with the highest similarity to merge, updating the hierarchy and the similarity between clusters after each merge, and repeating this process until a preset number of clusters is reached or the similarity between clusters is lower than a certain threshold.

[0128] In one embodiment of the present application, the screening module is used to construct an initial hierarchical structure based on a preset distance or similarity metric by calculating time and numerical variables, and form each order into an initial cluster separately to complete initialization; calculate the similarity between all cluster pairs, select the two cluster pairs with the highest similarity to merge, and after each merge, update the hierarchy and the similarity between clusters until the preset number of clusters is reached or the similarity between cluster pairs is lower than the preset similarity threshold.

[0129] In one embodiment of the present application, the loading constraint module is used to determine the basic loading constraints of the steel product order and the goods to be loaded, the basic loading constraints include that one steel product order can be loaded on at most one vehicle and one steel product order can be loaded in at most one hold; on the basis of the basic loading constraints, vehicle loading constraints and hold loading constraints are established, and the consistency of the goods on the ship and on the vehicle is ensured, the vehicle loading constraints include the same-vehicle loading principle, vehicle load constraints and vehicle loading quantity constraints, and the hold loading constraints include the same-hold loading principle, cargo area constraints and hold load constraints.

[0130] As an example, load constraints are imposed on the loadable resource pool based on the load reference attribute, including:

[0131] The same vehicle only carries the same cargo hold, that is, the same destination cargo;

[0132]

[0133] The same vehicle can only carry goods from the same cargo area;

[0134]

[0135] Only the same type of goods are loaded on the same vehicle;

[0136]

[0137] One order of steel products can be loaded on one vehicle at most;

[0138]

[0139] One order of steel products can be loaded in one hold at most;

[0140]

[0141] If steel products are loaded on a ship, they must also be loaded on a vehicle;

[0142]

[0143] The load of each cabin shall not exceed the upper load limit;

[0144]

[0145] The load of each cabin shall not be less than the lower load limit;

[0146]

[0147] The vehicle load does not exceed the upper limit;

[0148]

[0149] The vehicle load is not less than the lower limit;

[0150]

[0151] The vehicle load quantity does not exceed the upper limit;

[0152]

[0153] The vehicle loading quantity is not less than the lower limit;

[0154]

[0155] Among them, D is the destination arrival sequence set, I is the steel product order set in the available resource pool, T is the set of loading vehicles, C is the set of steel product varieties, S is the set of steel product cargo areas, and W is the set of steel product cargo areas. i is the weight attribute of steel product order i, N i is the quantity attribute of steel product order i, K ic K is the type attribute of steel product order i ic ∈{0,1},H id is the destination attribute H of steel product order i id ∈{0,1},J is is the storage area attribute J of steel product order i is ∈{0,1},D j is the arrival destination d of cabin j, SWU j is the upper limit of the loading constraint of cabin j, SWL j is the lower limit of the loading constraint of cabin j, TWU c The upper limit of the weight of steel type c loaded on vehicles, TWL c The minimum weight limit for loading steel type c on a vehicle, TNU c The upper limit of the quantity of steel type c loaded on the vehicle, TNL c W is the lower limit of the quantity of steel type c loaded on the vehicle, d is the expected loading weight at the destination, H id is the destination attribute H of steel product order i id ∈{0,1},a ij is a 0-1 variable for steel product i loaded in hold j, b it is a 0-1 variable for the steel product i loaded on vehicle t, e t Loading steel products into vehicle t, e t =0,if∑ i∈I b it =0,f ct Type of steel products loaded on vehicle t, f ct =1, if the vehicle is loaded with steel type c.

[0156] In one embodiment of the present application, the steel product water transport integrated loading device further includes a model building module 360 ​​for building a mixed integer programming model, including:

[0157]

[0158] Among them, Z is the constraint objective function; D is the destination arrival order set; I is the steel product order set in the available resource pool; T is the set of loading vehicles; W i is the weight attribute of steel product order i; W d is the expected loading weight at the destination; H id is the destination attribute of steel product order i, H id ∈{0,1}; a ij is a 0-1 variable for steel product i loaded in hold j; e t Loading steel products into vehicles; t =0,if∑ i∈I b it =0, b it is a 0-1 variable representing the steel product i loaded on vehicle t, and α and β are weight parameters.

[0159] In one embodiment of the present application, the solution module is used to set the search depth, the number of iterations and the neighborhood size of the solution, determine the basic framework of the search process, and obtain the algorithm parameters and constraints; according to the algorithm parameters and constraints, a first set of variables is set for the cargo hold loading condition of steel products to obtain the cargo hold loading initial solution, and a second set of variables is set for the cargo hold loading condition of steel products to obtain the vehicle loading initial solution; the cargo hold loading initial solution is coded and modified to obtain the cargo hold loading neighborhood, and the vehicle loading initial solution is coded and modified to obtain the vehicle loading neighborhood; the cargo hold loading initial solution is processed by deletion operator and insertion operator based on the cargo hold loading neighborhood, and the vehicle loading initial solution is processed by deletion operator and insertion operator based on the vehicle loading neighborhood to obtain the ship allocation details and the loading details.

[0160] In one embodiment of the present application, the solution module is used to randomly delete some cabin loading variables of the cabin loading initial solution within the cabin loading neighborhood to obtain the remaining cabin loading initial solution; insert the cabin loading operator into the remaining cabin loading initial solution to obtain the target cabin loading initial solution; and randomly delete some vehicle loading variables of the vehicle loading initial solution within the vehicle loading neighborhood to obtain the remaining vehicle loading initial solution; insert the vehicle loading operator into the remaining vehicle loading initial solution to obtain the target vehicle loading initial solution; integrate the target cabin loading initial solution and the target vehicle loading initial solution into the target loading initial solution; perform local search on the target loading initial solution, refine the target loading initial solution using a specific local optimization strategy, iteratively output the locally improved target loading initial solution, recorded as the iterative loading solution; evaluate the cabin loading efficiency of the iterative loading solution through the objective function, and output the current loading solution; during the iteration process, compare the objective function values ​​of the iterative loading solution and the current loading solution until the optimal loading solution is output; decode the optimal loading solution to obtain ship allocation details and loading details.

[0161] As an example, the specific steps to implement the large-scale neighborhood search algorithm are as follows:

[0162] Parameter initialization step: Set key algorithm parameters including but not limited to search depth, number of iterations, solution neighborhood size, etc. to establish the basic framework of the search process.

[0163] Generate initial solution step: Based on the preset parameters and constraints, construct a symbol coding system, which includes two types of 0-1 variables, representing the steel products loaded in the ship's hold and loaded in the vehicle. This step ensures that the position of each steel product in the loading plan is clearly mathematically represented. Specifically:

[0164] c. For the cargo hold loading, define a set of 0-1 variables, where each variable corresponds to the pairing of a specific steel product with a specific cargo hold. If the steel product is assigned to the cargo hold, the corresponding variable value is 1, otherwise it is 0.

[0165] d. For vehicle loading, a set of 0-1 variables is also defined, each variable corresponds to the pairing of a specific steel product with a specific vehicle. If the steel product is assigned to a vehicle, the variable value is 1, if not, it is 0.

[0166] Neighborhood definition step: The neighborhood solution set is formed by modifying the encoding of the initial solution. For steel products loaded in the hold and on the vehicle, the following definition is used to form the neighborhood:

[0167] c. Ship loading neighborhood: For each steel product loading variable in the ship, we consider moving the product from the current ship to another ship, or changing its loading status. This is achieved by changing the corresponding 0-1 encoding, that is, changing a variable from 0 to 1 to indicate loading, or from 1 to 0 to cancel loading.

[0168] d. Vehicle loading neighborhood: For the steel product loading variables on the vehicle, the possible transfer between products, between products and vehicles, or changing loading status are also considered. This is also achieved by adjusting the 0-1 variables corresponding to the steel products, modifying the variable values ​​to represent the different distribution states of the products on the vehicle.

[0169] Deletion operator and insertion operator application steps: By selecting or randomly determining specific variables for deletion operations, some parts of the initial solution are removed to explore new potential solutions. After deletion, the insertion operator is used to add the previously removed parts or new elements back to the solution, which requires evaluating their contribution to the objective function to determine the best insertion position.

[0170] c. Cabin loading: Randomly determine and delete specific cabin loading variables, set the 0-1 value to 0, indicating that the original allocation is cancelled. The insertion operator will consider reallocating or introducing new loading items into the cabin.

[0171] d. Vehicle loading: Randomly determine specific vehicle loading variables for deletion, and then explore optimized or new loading solutions through insertion operations.

[0172] The application steps of the local search algorithm are as follows: perform local search on the solutions remaining after the deletion operator is processed, refine the solution through a specific local optimization strategy, and seek local improvements of the solution.

[0173] Objective function evaluation step: The objective function is used to evaluate the performance of the solution obtained in each iteration, which includes a comprehensive evaluation of the loading rate of each flow direction of the ship and the number of vehicles used.

[0174] Solution selection and update steps: Compare the objective function values ​​of the iterative solution and the current optimal solution, and decide whether to use the iterative solution as the new current optimal solution based on the comparison result.

[0175] Termination condition determination step: Continue iterating until a pre-set termination condition is met, such as the quality of the solution no longer significantly improves, or the maximum number of iterations has been reached.

[0176] Output the optimal solution step: The optimal solution is finally selected. After decoding, the optimal solution obtains the ship allocation details and the loading details as the recommended solution for the integrated water transportation of finished steel products.

[0177] It should be noted that the device provided in the above embodiment and the method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In practical applications, the device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0178] See also Figure 3 , Figure 3 is a schematic diagram of a structure of an electronic device shown in an exemplary embodiment of the present application. It should be noted that: Figure 3 The electronic device 300 shown is only an example and should not bring any limitation to the functions and usage area of ​​the embodiments of the present application.

[0179] like Figure 3 As shown, the electronic device 300 includes a processor 301, a memory 302 and a communication bus 303; the communication bus 303 is used to connect the processor 301 and the memory 302; the processor 301 is used to execute the computer program stored in the memory 302 to implement one or more methods in the above embodiments.

[0180] The electronic device provided in the present application includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used to communicate, and the processor and the transceiver are used to run the computer program so that the electronic device executes each step of the above method.

[0181] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0182] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0183] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for integrated water transport loading of finished steel products, characterized in that: The method comprises: Obtaining finished steel product order information, ship information and basic stowage configuration information, wherein the ship information includes ship batch information, and the basic stowage configuration information includes information for constraining the finished steel product order; Marking key information in the finished steel product order information to obtain reference order information, and marking key attributes in the loading basic configuration information to obtain loading reference attributes; Based on the reference order information, a clustering algorithm model is used to filter the ship batch flow direction in the ship batch information to obtain a target ship batch flow direction, and an initial loading parameter is set for each target ship batch flow direction to obtain a ship batch flow direction screening result; The optimal reference order information cluster is selected from the screening results of the ship batch flow direction as a loadable resource pool, and the loadable resource pool is constrained based on the load reference attribute, with maximizing the loading rate of each ship flow direction and minimizing the number of vehicles used as the optimization goal, and a mixed integer programming model is established; The mixed integer programming model is solved by a large-scale neighborhood search algorithm to generate ship allocation details and vehicle loading details, thus completing the integrated water transportation loading of finished steel products.

2. The method for integrated water transportation of finished steel products according to claim 1, characterized in that: The clustering algorithm model is used to screen the ship batch flow direction in the ship batch information to obtain the target ship batch flow direction, including: Convert the time attribute in the reference order information into the number of days relative to the operation time of the clustering algorithm model, and record it as the operation time; encode the category attribute in the reference order information to obtain the encoding result, and record it as a numerical variable; According to the preset clustering requirements, the operation time and the numerical variables are hierarchically clustered to filter the ship batch flow direction in the ship batch information to obtain the target ship batch flow direction.

3. The method for integrated water transportation of finished steel products according to claim 2, characterized in that: Performing hierarchical clustering on the operation time and the numerical variable includes: According to a preset distance or similarity metric, an initial hierarchical structure is constructed through the operation time and the numerical variable, and each order is individually formed into an initial cluster to complete initialization; Calculate the similarity between all cluster pairs, select the two cluster pairs with the highest similarity to merge, and update the similarity between the hierarchy and clusters after each merge until the preset number of clusters is reached or the similarity between cluster pairs is lower than the preset similarity threshold.

4. The method for integrated water transportation of finished steel products according to claim 1, characterized in that: The load-bearing resource pool is constrained based on the load-bearing reference attribute, including: Determine basic loading constraints for steel product orders and goods to be loaded, wherein the basic loading constraints include that one steel product order can be loaded on at most one vehicle and one steel product order can be loaded in at most one ship cabin; On the basis of the basic loading constraints, vehicle loading constraints and cabin loading constraints are established to ensure the consistency of cargo on board and on the vehicle. The vehicle loading constraints include the same-vehicle loading principle, vehicle load constraints and vehicle loading quantity constraints. The cabin loading constraints include the same-cabin loading principle, cargo area constraints and cabin load constraints.

5. The method for integrated water transportation of finished steel products according to claim 1, characterized in that: With the optimization objectives of maximizing the loading rate of ships in each flow direction and minimizing the number of vehicles used, a mixed integer programming model is established, including: Among them, Z is the constraint objective function; D is the destination arrival order set; I is the set of steel product orders in the available resource pool; T is the set of loading vehicles; W i is the weight attribute of steel product order i; W d The desired load weight for the destination; H id is the destination attribute of steel product order i, H id ∈{0,1}; a ij is a 0-1 variable for steel product i loaded in hold j; e t Loading steel products onto vehicles; e t =0,if∑ i∈I b it =0, b it is a 0-1 variable representing the steel product i loaded on vehicle t, and α and β are weight parameters.

6. The method for integrated water transportation of finished steel products according to claim 1, characterized in that: The designed mixed integer programming model is solved by a large-scale neighborhood search algorithm to generate ship allocation details and loading details, including: Set the search depth, number of iterations, and solution neighborhood size, determine the basic framework of the search process, and obtain algorithm parameters and constraints; According to the algorithm parameters and constraints, a first set of variables is set for the loading of steel products in a ship cabin to obtain an initial solution for loading of the ship cabin, and a second set of variables is set for the loading of steel products in a vehicle to obtain an initial solution for loading of the vehicle; Modifying the initial solution of cabin loading by coding to obtain a cabin loading neighborhood, and modifying the initial solution of vehicle loading by coding to obtain a vehicle loading neighborhood; The initial solution of the cabin loading is processed by a deletion operator and an insertion operator based on the cabin loading neighborhood, and the initial solution of the vehicle loading is processed by a deletion operator and an insertion operator based on the vehicle loading neighborhood to obtain ship allocation details and vehicle loading details.

7. The method for integrated water transportation of finished steel products according to claim 6, characterized in that: The initial solution of the cabin loading is processed by a deletion operator and an insertion operator based on the cabin loading neighborhood, and the initial solution of the vehicle loading is processed by a deletion operator and an insertion operator based on the vehicle loading neighborhood to obtain ship allocation details and vehicle loading details, including: Randomly delete some cabin loading variables of the cabin loading initial solution in the cabin loading neighborhood to obtain a remaining cabin loading initial solution; insert cabin loading operators into the remaining cabin loading initial solution to obtain a target cabin loading initial solution; and randomly delete some vehicle loading variables of the vehicle loading initial solution in the vehicle loading neighborhood to obtain a remaining vehicle loading initial solution; insert vehicle loading operators into the remaining vehicle loading initial solution to obtain a target vehicle loading initial solution; Integrating the target cabin loading initial solution and the target vehicle loading initial solution into a target loading initial solution; Performing a local search on the target loading initial solution, refining the target loading initial solution using a specific local optimization strategy, and iteratively outputting a locally improved target loading initial solution, which is recorded as an iterative loading solution; Performing cabin loading efficiency evaluation on the iterative loading solution through an objective function, and outputting a current loading solution; During the iteration process, the objective function value of the iterative loading solution is compared with the objective function value of the current loading solution until the optimal loading solution is output; The optimal loading solution is decoded to obtain ship allocation details and vehicle loading details.

8. An integrated loading device for water transport of finished steel products, characterized in that: The device comprises: An acquisition module, used to acquire finished steel product order information, ship information and basic stowage configuration information, wherein the ship information includes ship batch information, and the basic stowage configuration information includes information for constraining finished steel product order information; A marking module, used to mark key information in the steel finished product order information to obtain reference order information, and mark key attributes in the loading basic configuration information to obtain loading reference attributes; A screening module is used to screen the ship batch flow direction in the ship batch information based on the reference order information by using a clustering algorithm model to obtain a target ship batch flow direction, set initial loading parameters for each target ship batch flow direction, and obtain a ship batch flow direction screening result; A loading constraint module is used to select the optimal reference order information cluster from the ship batch flow direction screening result as the loadable resource pool, and to perform loading constraints on the loadable resource pool based on the loading reference attribute, so as to maximize the loading rate of each ship flow direction and minimize the number of vehicles used as the optimization goal, and establish a mixed integer programming model; The solution module is used to solve the mixed integer programming model through a large-scale neighborhood search algorithm, generate ship allocation details and vehicle loading details, and complete the integrated water transportation of finished steel products.

9. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the integrated water transportation loading method for finished steel products as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to enable a computer to execute the integrated water transportation loading method for finished steel products as described in any one of claims 1 to 7.

Images