Two-dimensional loading optimization method, device, computer equipment and storage medium

Abstract

The present application relates to a two-dimensional loading optimization method, device, computer equipment and storage medium. By obtaining the size information of the items to be loaded; obtaining an initial loading plan based on the size information and preset freight vehicle information; optimizing the initial loading plan through a preset column generation model to obtain a two-dimensional loading plan. The present application first obtains the size information of the items to be loaded, and then generates a two-dimensional loading plan based on the initial loading plan under a loop framework of column generation by using a column generation model. Based on the model decomposition effect of the column generation algorithm, the scale of the processing model of the loading optimization process can be significantly reduced, the model solving efficiency can be improved, and thus the processing efficiency of the loading optimization process can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a two-dimensional loading optimization method, device, computer equipment and storage medium. Background Art

[0002] With the development of Internet technology, online shopping has become a lifestyle for many people. The logistics industry has also developed rapidly. In the logistics scene of large-scale delivery in the logistics industry, due to the large size of express parcels, when the volume reaches a certain level, it may not be possible to transport them all at once by one vehicle. At this time, multiple vehicles or the same vehicle is required to load and transport them multiple times. Based on this situation, loading efficiency (i.e. the space utilization rate of loading) is crucial to saving delivery costs.

[0003] For specific large-item loading scenarios, since large items are heavy, they are not allowed to be stacked on top of each other to avoid crushing the items, that is, all items can only be placed in one layer. In this way, the original three-dimensional space in the car is compressed into two dimensions, and the maximum use of the space in the car is converted into the maximum occupation of the bottom area of ​​the car.

[0004] However, when the number of express parcels is large, the total number of vehicles required is difficult to determine, and it is difficult to solve using the conventional primitive planning model, which affects the loading efficiency. Summary of the invention

[0005] Based on this, it is necessary to provide a two-dimensional loading optimization method, device, computer equipment and storage medium to improve the loading efficiency in the case of large-item delivery in response to the above-mentioned technical problems.

[0006] A two-dimensional loading optimization method, the method comprising:

[0007] Get the size information of the items to be loaded;

[0008] Obtaining an initial loading plan based on the size information and preset freight vehicle information;

[0009] The initial loading plan is optimized by a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate a loading alternative plan based on the initial loading plan, and to find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0010] In one embodiment, obtaining an initial loading plan according to the size information and preset freight vehicle information includes:

[0011] Determining the size type of the object to be loaded according to the size information of the object to be loaded;

[0012] According to the preset freight vehicle information, construct a loading plan corresponding to the items to be loaded of each size type;

[0013] Determine the initial loading plan based on the loading plans corresponding to the items to be loaded of each size type.

[0014] In one embodiment, determining the size type of the object to be loaded according to the size information of the object to be loaded includes:

[0015] Get the size information of the items to be loaded;

[0016] Obtaining the similarity of size information between the objects to be loaded;

[0017] Classifying the objects to be loaded according to the similarity of the size information;

[0018] Get the size type corresponding to the item to be loaded.

[0019] In one embodiment, the optimizing the initial loading plan by using the preset column generation model to obtain the two-dimensional loading plan also includes:

[0020] With the goal of minimizing the number of vehicles used, a constrained master problem of the preset column generation model is constructed;

[0021] Constructing the dual problem of the master problem of the preset column generation model according to the restricted master problem;

[0022] Based on the preset two-dimensional loading basic constraints and the constraints of the dual problem of the main problem, a price sub-problem of the preset column generation model is constructed;

[0023] A preset column generation model is constructed according to the restriction main problem, the dual problem of the main problem and the price sub-problem.

[0024] In one embodiment, the price sub-problem of constructing the preset column generation model based on the preset two-dimensional loading basic constraints and the constraints of the dual problem of the main problem includes:

[0025] Constructing constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, wherein the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints, and maximum loading area constraints;

[0026] Constructing the objective function of the price sub-problem by maximally violating the constraints of the dual problem of the main problem;

[0027] A price sub-problem is constructed according to the preset model parameters, the preset decision variables, the constraint conditions and the objective function.

[0028] In one embodiment, optimizing the initial loading plan by using a preset column generation model to obtain a two-dimensional loading plan includes:

[0029] Obtain the optimization result of the restricted main problem corresponding to the initial loading plan

[0030] Obtaining the optimal solution of the dual problem of the master problem in the preset column generation model according to the optimization result of the restricted master problem;

[0031] According to the optimal solution, generating a loading alternative plan by presetting the price sub-problem in the column generation model;

[0032] A two-dimensional loading plan is obtained according to the loading alternative plan.

[0033] In one embodiment, after optimizing the initial loading plan by using the preset column generation model to obtain the two-dimensional loading plan, the method further includes:

[0034] The two-dimensional loading scheme is optimized by using a preset main heuristic model to obtain an optimized loading scheme;

[0035] Feedback on the loading optimization plan.

[0036] A two-dimensional loading optimization device, comprising:

[0037] A data acquisition module is used to obtain the size information of the items to be loaded;

[0038] An initial plan acquisition module, used to acquire an initial loading plan according to the size information and preset freight vehicle information;

[0039] A scheme optimization module is used to optimize the initial loading scheme through a preset column generation model to obtain a two-dimensional loading scheme. The preset column generation model is used to generate a loading alternative scheme based on the initial loading scheme, and to find a two-dimensional loading scheme that uses the least number of vehicles from the loading alternative schemes.

[0040] In one of the embodiments, the initial plan acquisition module is specifically used to: determine the size type of the items to be loaded according to the size information of the items to be loaded; construct loading plans corresponding to the items to be loaded of each size type according to the preset freight vehicle information; and determine the initial loading plan according to the loading plans corresponding to the items to be loaded of each size type.

[0041] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0042] Get the size information of the items to be loaded;

[0043] Obtaining an initial loading plan based on the size information and preset freight vehicle information;

[0044] The initial loading plan is optimized by a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate a loading alternative plan based on the initial loading plan, and to find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0045] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0046] Get the size information of the items to be loaded;

[0047] Obtaining an initial loading plan based on the size information and preset freight vehicle information;

[0048] The initial loading plan is optimized by a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate a loading alternative plan based on the initial loading plan, and to find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0049] The above-mentioned two-dimensional loading optimization method, device, computer equipment and storage medium obtain the size information of the items to be loaded; obtain the initial loading plan according to the size information and the preset freight vehicle information; optimize the initial loading plan by the preset column generation model to obtain the two-dimensional loading plan. The present application first obtains the size information of the items to be loaded, and then generates a two-dimensional loading plan based on the initial loading plan under the loop framework of column generation by using the column generation model. Based on the model decomposition effect of the column generation algorithm, the processing model scale of the loading optimization process can be significantly reduced, the model solving efficiency can be improved, and then the processing efficiency of the loading optimization process can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A diagram showing an application environment of a two-dimensional loading optimization method in an embodiment;

[0051] Figure 2 A schematic diagram of a flow chart of a two-dimensional loading optimization method in one embodiment;

[0052] Figure 3 In one embodiment Figure 2 Schematic diagram of the sub-process of step 201;

[0053] Figure 4 In one embodiment Figure 2 Schematic diagram of the sub-process of step 205;

[0054] Figure 5 In another embodiment Figure 2 Schematic diagram of the sub-process of step 205;

[0055] Figure 6 In one embodiment Figure 2 Schematic diagram of the sub-process of step 207;

[0056] Figure 7 It is a structural block diagram of a two-dimensional loading optimization device in one embodiment;

[0057] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] The two-dimensional loading optimization method provided in this application can be applied to Figure 1 In the application environment shown. Among them, the work terminal 102 where the loading staff is located communicates with the two-dimensional loading optimization server 104 through the network. When it is necessary to plan large-scale loading, the work terminal 102 can send the size information of the items to be loaded to the two-dimensional loading optimization server 104, and then the two-dimensional loading optimization server 104 obtains the size information of the items to be loaded; according to the size information and the preset freight vehicle information, the initial loading plan is obtained; the initial loading plan is optimized by the preset column generation model to obtain a two-dimensional loading plan, and the preset column generation model is used to generate a loading alternative plan according to the initial loading plan, and find the two-dimensional loading plan with the least number of vehicles from the loading alternative plan. Among them, the work terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the two-dimensional loading optimization server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0060] In one embodiment, Figure 2 As shown in the figure, a two-dimensional loading optimization method is provided, which is applied to Figure 1 The two-dimensional loading optimization server 104 in the example is used for explanation, and the following steps are included:

[0061] Step 201, obtaining the size information of the items to be loaded.

[0062] Among them, the two-dimensional loading optimization method of the present application is specifically used to optimize the process of loading large items in logistics. In the logistics scene of large-scale delivery in the logistics industry, due to the large volume of express parcels, when the amount of parcels reaches a certain level, it may not be possible to complete the delivery by one vehicle at a time. At this time, multiple vehicles or the same vehicle are required to load and transport multiple times. Based on this situation, loading efficiency is crucial to saving delivery costs. Due to the large weight of large items, they are not allowed to be stacked in an overlapping manner to avoid crushing the express parcels, that is, all express parcels can only be placed in one layer. At this time, the loading problem can be specifically regarded as a two-dimensional loading optimization problem, and the maximum use of the space in the car is converted into the maximum occupation of the bottom area in the car. Therefore, the size information of the items to be loaded at this time specifically refers to the three-dimensional information of the length, width and height of the items. That is, the length, width and height of the items after they are packaged into express parcels. Large-scale loading does not allow overlapping placement, so the three-dimensional model is converted into two dimensions, which reduces the complexity of the problem to a certain extent, but an additional problem needs to be considered, that is, the placement of express parcels. In a three-dimensional model, six placement methods need to be considered; in the two-dimensional model of this article, the bottom surface of the express can be determined first, and then the horizontal and vertical placement methods of the bottom surface can be considered. For express parcels with special requirements, that is, one side must be the bottom surface, the bottom surface is determined; for express parcels without special requirements, the smallest possible surface is selected as the bottom surface, as long as its vertical height does not exceed the height of the compartment, so as to ensure that the minimum area is occupied, and more express parcels can be loaded.

[0063] Specifically, in the scenario of loading large items, the staff at the loading site needs to determine whether to perform two-dimensional loading optimization based on actual needs. When the number of items to be loaded is large and multiple delivery vehicles are required for delivery, or when multiple batches of transportation are required, the two-dimensional loading optimization method of this application can be used to optimize the placement of express items in the car, thereby matching the minimum number of vehicles for loading express items, which can also be regarded as the minimum number of round trips for delivery vehicles. This saves the transportation cost and manpower cost of the express transportation process and improves the efficiency of express delivery. When the staff at the loading site encounters large items during loading and needs to be transported multiple times, the size information of the items to be loaded can be directly submitted to the two-dimensional loading optimization server 104, and the two-dimensional loading optimization server 104 will optimize the loading process.

[0064] Step 203, obtaining an initial loading plan based on the size information and preset freight vehicle information.

[0065] Among them, the preset freight vehicle information refers to the internal information of the vehicle used to transport the items to be loaded. Specifically includes information such as the length, width, and height of the freight compartment in the freight vehicle. In one embodiment, this part of information can be pre-stored in the two-dimensional loading optimization server 104. When submitting the size information of the item, the work terminal 102 can simultaneously push the vehicle type information used to transport the items to be loaded, and then the two-dimensional loading optimization server 104 can effectively search for the preset freight vehicle information corresponding to the vehicle type information in the preset database based on this part of the vehicle type information. The initial loading plan is the initial plan for optimizing the two-dimensional loading plan, and is the optimization target of the preset column generation model. In this application, the two-dimensional loading optimization is solved by a column generation algorithm. Since the column generation algorithm requires a set of initial feasible solutions as initial alternatives, it is necessary to design a heuristic method for obtaining the initial plan.

[0066] Specifically, the solution of the present application optimizes the two-dimensional loading process through a column generation model. Before performing the column generation optimization, it is necessary to determine the initial optimization plan to serve as the basis for optimization. At this time, the initial loading plan can be obtained based on the size information and the preset freight vehicle information. Specifically, in one of the specific embodiments, for each size type of express, there is a loading method that only loads this type of express (maximizing the loading quantity), so there are as many initial alternative solutions as there are size types, and since each size type has a corresponding solution, the needs of each size type can be covered, and the initial solution must be feasible.

[0067] Step 205, optimize the initial loading plan through a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate a loading alternative plan based on the initial loading plan, and find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0068] Among them, the column generation model is a mathematical model built on the basis of the column generation algorithm, and the column generation algorithm is a very efficient algorithm for solving large-scale linear optimization problems. Specifically, this application takes the obtained initial loading plan as the object of optimization. Due to the large number of express parcels, the total number of vehicles required is difficult to determine, and it is difficult to solve using the conventional original planning model. Therefore, it is considered to use the column generation model to solve the number of alternative solutions used in the main problem, generate new alternative solutions in the sub-problem, and continuously generate alternative solutions that may provide optimization through the loop framework of column generation, so as to obtain a more optimized two-dimensional loading plan.

[0069] The above two-dimensional loading optimization method obtains the size information of the items to be loaded; obtains the initial loading plan based on the size information and the preset freight vehicle information; optimizes the initial loading plan through the preset column generation model to obtain the two-dimensional loading plan. The present application first obtains the size information of the items to be loaded, and then generates a two-dimensional loading plan based on the initial loading plan under the loop framework of column generation by using the column generation model. Based on the model decomposition effect of the column generation algorithm, the processing model scale of the loading optimization process can be significantly reduced, the model solving efficiency can be improved, and then the processing efficiency of the loading optimization process can be improved.

[0070] In one embodiment, Figure 3 As shown, step 201 includes:

[0071] Step 302: Determine the size type of the object to be loaded according to the size information of the object to be loaded.

[0072] Step 304: construct a loading plan corresponding to the items to be loaded of each size type according to the preset freight vehicle information.

[0073] Step 306: Determine an initial loading plan according to the loading plans corresponding to the items to be loaded of each size type.

[0074] The size type is used to summarize different items to be loaded. Items to be loaded with similar length, width and height can be regarded as the same size type. Then, the initial loading plan is constructed, and the subsequent loading optimization is processed, thereby reducing the complexity of the solution process.

[0075] Specifically, in the process of designing the initial loading plan, for each size type of express, there is a loading method that only loads this type of express (maximizing the loading quantity), so there are as many initial alternative plans as there are size types, and since each size type has a corresponding plan, the needs of each size type can be covered, and the initial solution must be feasible. In this embodiment, by determining the size type, the loading plan corresponding to the items to be loaded of all size types can be effectively constructed, and then the initial loading plan can be efficiently determined, thereby improving the efficiency of constructing the initial loading plan.

[0076] like Figure 4 As shown, in one embodiment, step 302 includes:

[0077] Step 401, obtaining the size information of the items to be loaded.

[0078] Step 403, obtaining the similarity of the size information between the items to be loaded.

[0079] Step 405: Classify the items to be loaded according to the similarity of the size information.

[0080] Step 407, obtaining the size type corresponding to the items to be loaded.

[0081] The similarity is used to reflect the similarity of the length, width, height and other information between different items to be loaded. Different items to be loaded with similar sizes can be considered to be of the same size type.

[0082] Specifically, in the process of determining the loading type, the size information of the items to be loaded can be obtained first, including the three types of information of length, width and height. In this process, the longest, medium and shortest information in the three-dimensional information of the items can be named as the length, width and height in the size information, so that a three-dimensional array (length, width, height) corresponding to the items to be loaded is obtained for representing the size information. Therefore, the process of obtaining the similarity of the size information between the items to be loaded becomes a problem of comparing the similarity between the three-dimensional arrays corresponding to the size information. In one embodiment, the angle between the corresponding vectors of the three-dimensional array can be calculated, and the size information with an angle less than a preset angle threshold is regarded as similar size information. Then, the items to be loaded are classified according to the similar size information. In one embodiment, the calculation results of the similarity calculation process can be clustered, and then the items to be loaded are classified according to the clustering results. In this embodiment, by obtaining the similarity of the size information between the items to be loaded to perform the initial classification of the items to be loaded, the planning effectiveness of the initial loading plan planning process can be effectively guaranteed, thereby improving the efficiency and accuracy of loading optimization.

[0083] like Figure 5 As shown, in one embodiment, before step 205, the process further includes:

[0084] Step 502, with the goal of minimizing the number of vehicles used, construct a constrained main problem of a preset column generation model.

[0085] Step 504, constructing the dual problem of the master problem of the preset column generation model according to the restricted master problem.

[0086] Step 506, based on the preset two-dimensional loading basic constraints and the constraints of the dual problem of the main problem, construct the price sub-problem of the preset column generation model.

[0087] Step 508, constructing a preset column generation model according to the restriction main problem, the dual problem of the main problem and the price sub-problem.

[0088] In one embodiment, step 506 includes: constructing constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints and maximum loading area constraints; constructing the objective function of the price sub-problem with the constraints that violate the dual problem of the main problem to the greatest extent; constructing the price sub-problem according to preset model parameters, preset decision variables, constraints and objective function.

[0089] Among them, the restricted main problem, the dual problem of the main problem and the price subproblem are all part of the column generation algorithm framework. By constructing the restricted main problem, the dual problem of the main problem and the price subproblem, a preset column generation model for two-dimensional loading optimization can be effectively generated. The restricted main problem is the main problem of the column generation algorithm because all alternative solutions are restricted to the generated solution set. Before entering the column generation loop framework, the solution set only has the initial solution generated by the heuristic method; after entering the column generation framework, each time the subproblem is solved, a new solution is generated. If the optimal solution of the subproblem meets the optimization conditions, the new solution is added to the solution set, otherwise the loop ends and the solution set is no longer expanded. The optimization goal of the main problem is the same as the overall goal, that is, to minimize the number of alternative solutions used (minimize the number of vehicles), but this problem is a relaxation problem (the variables are continuous variables rather than integer variables), and the purpose is to obtain the optimal solution of the dual problem to determine the optimization goal of the subproblem. In one embodiment, the restricted main problem specifically includes four parts: model parameters, decision variable constraints and objective function. In one embodiment, the model parameters include:

[0090] I A collection of all express size types

[0091] The set of all K solutions

[0092] D i Number of shipments of size type i

[0093] a k,i The number of shipments of size type i in solution k

[0094] The decision variables include:

[0095] x k The number of schemes k used

[0096] The constraints include:

[0097] The loading requirements of each size type of express shipment are met

[0098]

[0099] The objective function includes:

[0100] Minimize the number of alternatives used, that is, minimize the number of vehicles.

[0101]

[0102] Correspondingly, the model structure of the dual problem of the main problem is:

[0103] Decision variables:

[0104] y i (The decision variables of the dual problem correspond to the constraints of the original problem)

[0105] Constraints:

[0106]

[0107] Objective function:

[0108]

[0109] The subproblem is the price subproblem. The so-called "price" refers to the degree of violation of the constraints of the dual problem of the main problem. Since the alternative solutions for the main problem are limited (a subset of all theoretically existing solutions), these alternative solutions may not necessarily produce good optimization results. Therefore, the subproblem needs to generate new alternative solutions. The generation principle is whether the constraints of the dual problem of the main problem can be violated, and the solution with the strongest degree of violation is selected. Therefore, this is the optimization goal of the subproblem; and the constraints of the subproblem are the basic constraints of the two-dimensional loading problem. In one embodiment, the specific model structure of the subproblem model is as follows:

[0110] Model parameters: (The parameters mentioned above will not be repeated)

[0111] N i The maximum number of type i shipments that can be loaded on a vehicle

[0112] X The width of the vehicle from side to side

[0113] Y The length of the vehicle in the inner and outer directions

[0114] w i The length of the long side of a type i shipment

[0115] h i Short side length of type i shipment

[0116] The Optimal Solution to the Dual Problem of the Main Problem

[0117] Decision variables:

[0118]

[0119]

[0120] Constraints:

[0121] 1) There is only one placement method:

[0122]

[0123] 2) If two parcels are loaded, there is at least one relationship between their left-right and inside-outside positions:

[0124]

[0125]

[0126] 3) Vehicle size:

[0127]

[0128]

[0129] 4) If two parcels are loaded, the coordinate size relationship must satisfy the position relationship in the left-right and inside-out directions:

[0130]

[0131]

[0132]

[0133] 5) Get the number of shipments of each size type:

[0134]

[0135] 6) Effective inequality, maximum loading area (used to speed up the solution of sub-problems):

[0136]

[0137] Objective function:

[0138] The constraints of the dual problem of the main problem are violated to the greatest extent.

[0139]

[0140] In this embodiment, by constructing a restricted main problem, a dual problem of the main problem, and a price sub-problem, a preset train generation model for loading optimization is effectively constructed to ensure the accuracy of the train generation model optimization.

[0141] like Figure 6 As shown, in one embodiment, step 205 includes:

[0142] Step 601, obtaining the optimization result of the constrained main problem corresponding to the initial loading plan.

[0143] Step 603, obtaining the optimal solution of the dual problem of the main problem in the preset column generation model according to the optimization result of the restricted main problem.

[0144] Step 605, based on the optimal solution, generate loading alternatives by presetting the price sub-problem in the column generation model.

[0145] Step 607, obtaining a two-dimensional loading plan according to the loading alternative plans, the preset column generation model is used to generate the loading alternative plans according to the initial loading plan, and to find the two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0146] Specifically, in the process of solving the two-dimensional loading scheme, the initial optimization is first performed based on the initial loading scheme and the restricted main problem in the model to obtain the corresponding optimization results. The optimization results of the restricted main problem are mainly used to optimize the number of alternative solutions, that is, to minimize the number of vehicles (loading times) required for the loading process. Then the optimal solution of the dual problem of the main problem is obtained to determine the optimization target of the price subproblem, so as to continuously generate new loading alternatives through the price subproblem. After obtaining the new loading alternatives, when the optimal solution of the subproblem is less than or equal to 0, the loading alternative obtained by the subproblem is used as the initial loading scheme, and the step of obtaining the optimization results of the restricted main problem corresponding to the initial loading scheme is returned to the loop processing. If the solution obtained by the price subproblem can no longer violate the constraints of the dual problem of the main problem (that is, the optimal solution of the subproblem is not less than 0), it is considered that all potential solutions that can optimize the main problem target have been found, so the loop ends. In this embodiment, the initial loading scheme is optimized by the preset column generation model, and effective loading optimization can be performed when the number of express parcels is large and the total number of vehicles required is difficult to determine, forming an available loading scheme to ensure the effectiveness of loading.

[0147] In one of the embodiments, after step 205, the process further includes: optimizing the two-dimensional loading plan by using a preset main heuristic model to obtain an optimized loading plan; and providing feedback on the optimized loading plan.

[0148] Specifically, when the solution of the subproblem in the preset column generation model can no longer violate the constraint conditions of the dual problem of the main problem (that is, the optimal solution of the subproblem is not less than 0), it is considered that all potential solutions that can optimize the main problem target have been found, and the processing loop of the preset column generation model can be ended. Since the main problem in the column generation loop is a relaxed problem, and the branch and bound operation in integer programming is not performed, the loop for generating new solutions may be incompletely generated. At this time, it can be processed by the preset main heuristic model, and the final processing optimization has been performed. The model structure of the preset main heuristic model is basically the same as that of the restricted main problem, with the following differences: the decision variable is changed from a continuous variable to an integer variable; at the same time, the set of all solutions is expanded due to the operation of the column generation algorithm. Through the final optimization of the main heuristic model, the usefulness of the obtained loading optimization solution can be effectively guaranteed, and the loading optimization solution can be effectively implemented.

[0149] It should be understood that although Figure 2-6 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-6 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0150] In one embodiment, Figure 7 As shown, a two-dimensional loading optimization device is provided, comprising:

[0151] The data acquisition module 702 is used to acquire the size information of the items to be loaded.

[0152] The initial solution acquisition module 704 is used to acquire an initial loading solution according to the size information and the preset freight vehicle information.

[0153] The scheme optimization module 706 is used to optimize the initial loading scheme through a preset column generation model to obtain a two-dimensional loading scheme. The preset column generation model is used to generate a loading alternative scheme based on the initial loading scheme, and to find a two-dimensional loading scheme that uses the least number of vehicles from the loading alternative schemes.

[0154] In one of the embodiments, the initial plan acquisition module is specifically used to: determine the size type of the items to be loaded according to the size information of the items to be loaded; construct loading plans corresponding to the items to be loaded of each size type according to the preset freight vehicle information; and determine the initial loading plan according to the loading plans corresponding to the items to be loaded of each size type.

[0155] In one embodiment, the initial solution acquisition module is further used to: obtain size information of the items to be loaded; obtain the similarity of size information between the items to be loaded; classify the items to be loaded according to the similarity of size information; and obtain the size type corresponding to the items to be loaded.

[0156] In one of the embodiments, it also includes a model construction module, which is used to: construct a restricted main problem of the preset column generation model with the goal of minimizing the number of vehicles used; construct the main problem dual problem of the preset column generation model based on the restricted main problem; construct a price sub-problem of the preset column generation model based on the preset two-dimensional loading basic constraints and the constraints of the main problem dual problem; construct the preset column generation model based on the restricted main problem, the main problem dual problem and the price sub-problem.

[0157] In one embodiment, the model construction module is also used to: construct constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints and maximum loading area constraints; construct the objective function of the price sub-problem with the constraints that violate the dual problem of the main problem to the greatest extent; construct the price sub-problem according to preset model parameters, preset decision variables, constraints and objective function.

[0158] In one embodiment, the solution optimization module 706 is specifically used to: obtain the optimization result of the restricted main problem corresponding to the initial loading plan; obtain the optimal solution of the dual problem of the main problem in the preset column generation model according to the optimization result of the restricted main problem; generate a loading alternative plan through the price sub-problem in the preset column generation model according to the optimal solution; and obtain a two-dimensional loading plan according to the loading alternative plan.

[0159] In one of the embodiments, a post-processing module is further included, which is used to: optimize the two-dimensional loading plan through a preset main heuristic model to obtain an optimized loading plan; and feed back the optimized loading plan.

[0160] The specific definition of the two-dimensional loading optimization device can be found in the definition of the two-dimensional loading optimization method above, which will not be repeated here. Each module in the above-mentioned two-dimensional loading optimization device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0161] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store information recommendation data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a two-dimensional loading optimization method is implemented.

[0162] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0163] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0164] Get the size information of the items to be loaded;

[0165] Obtain the initial loading plan based on the size information and preset freight vehicle information;

[0166] The initial loading plan is optimized by a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate loading alternative plans based on the initial loading plan, and to find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

[0167] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the size type of the items to be loaded based on the size information of the items to be loaded; constructing loading plans corresponding to the items to be loaded of each size type based on the preset freight vehicle information; and determining an initial loading plan based on the loading plans corresponding to the items to be loaded of each size type.

[0168] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining size information of the items to be loaded; obtaining the similarity of the size information between the items to be loaded; classifying the items to be loaded according to the similarity of the size information; and obtaining the size type corresponding to the items to be loaded.

[0169] In one embodiment, when the processor executes the computer program, the following steps are also implemented: constructing a restricted main problem of a preset column generation model with the goal of minimizing the number of vehicles used; constructing a dual problem of the main problem of the preset column generation model based on the restricted main problem; constructing a price subproblem of the preset column generation model based on the preset two-dimensional loading basic constraints and the constraints of the dual problem of the main problem; constructing a preset column generation model based on the restricted main problem, the dual problem of the main problem and the price subproblem.

[0170] In one embodiment, when the processor executes the computer program, the following steps are also implemented: constructing constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints and maximum loading area constraints; constructing the objective function of the price sub-problem with the constraints that violate the dual problem of the main problem to the greatest extent; constructing the price sub-problem according to preset model parameters, preset decision variables, constraint conditions and objective function.

[0171] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the optimization result of the restricted main problem corresponding to the initial loading plan; obtaining the optimal solution of the dual problem of the main problem in the preset column generation model based on the optimization result of the restricted main problem; generating a loading alternative plan through the price sub-problem in the preset column generation model based on the optimal solution; and obtaining a two-dimensional loading plan based on the loading alternative plan.

[0172] In one embodiment, when the processor executes the computer program, the following steps are also implemented: optimizing the two-dimensional loading plan by using a preset main heuristic model to obtain an optimized loading plan; and feeding back the optimized loading plan.

[0173] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0174] Get the size information of the items to be loaded;

[0175] Obtain an initial loading plan according to the size information and the preset freight vehicle information;

[0176] Optimize the initial loading plan through a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate alternative loading plans according to the initial loading plan and find the two-dimensional loading plan with the fewest number of vehicles used from the alternative loading plans.

[0177] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: determine the size type of the items to be loaded according to the size information of the items to be loaded; construct a loading plan corresponding to the items to be loaded of each size type according to the preset freight vehicle information; determine the initial loading plan according to the loading plans corresponding to the items to be loaded of each size type.

[0178] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: obtain the size information of the items to be loaded; obtain the similarity of the size information between the items to be loaded; classify the items to be loaded according to the similarity of the size information; obtain the corresponding size type of the items to be loaded.

[0179] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: construct a restricted master problem of the preset column generation model with the goal of minimizing the number of vehicles used; construct the dual problem of the master problem of the preset column generation model according to the restricted master problem; construct a price sub-problem of the preset column generation model based on the preset two-dimensional loading basic constraints and the constraints of the dual problem of the master problem; construct the preset column generation model according to the restricted master problem, the dual problem of the master problem, and the price sub-problem.

[0180] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: construct the constraint conditions of the price sub-problem according to the preset two-dimensional loading basic constraints, and the preset two-dimensional loading basic constraints include placement method constraints, direction and position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints, and maximum loading area constraints; construct the objective function of the price sub-problem by maximizing the violation of the constraints of the dual problem of the master problem; construct the price sub-problem according to the preset model parameters, preset decision variables, constraint conditions, and objective function.

[0181] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: obtain the optimization result of the restricted master problem corresponding to the initial loading plan; obtain the optimal solution of the dual problem of the master problem in the preset column generation model according to the optimization result of the restricted master problem; generate alternative loading plans through the price sub-problem in the preset column generation model according to the optimal solution; obtain the two-dimensional loading plan according to the alternative loading plans.

[0182] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: optimizing the two-dimensional loading plan by using a preset main heuristic model to obtain an optimized loading plan; and feeding back the optimized loading plan.

[0183] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0184] The technical features of the above embodiments may be combined arbitrarily. 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.

[0185] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A two-dimensional loading optimization method, the method include: Get the size information of the items to be loaded; Obtaining an initial loading plan based on the size information and preset freight vehicle information; With the goal of minimizing the number of vehicles used, a constrained master problem of the preset column generation model is constructed; Constructing the dual problem of the master problem of the preset column generation model according to the restricted master problem; Constructing constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, wherein the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints, and maximum loading area constraints; Constructing the objective function of the price sub-problem by maximally violating the constraints of the dual problem of the main problem; Constructing a price sub-problem according to preset model parameters, preset decision variables, the constraint conditions and the objective function; Constructing a preset column generation model according to the restriction main problem, the dual problem of the main problem and the price subproblem; The initial loading plan is optimized by a preset column generation model to obtain a two-dimensional loading plan. The preset column generation model is used to generate a loading alternative plan based on the initial loading plan, and to find a two-dimensional loading plan that uses the least number of vehicles from the loading alternative plans.

2. The method according to claim 1, It is characterized in that The obtaining of the initial loading plan according to the size information and the preset freight vehicle information includes: Determining the size type of the object to be loaded according to the size information of the object to be loaded; According to the preset freight vehicle information, construct a loading plan corresponding to the items to be loaded of each size type; Determine the initial loading plan based on the loading plans corresponding to the items to be loaded of each size type.

3. The method according to claim 2, It is characterized in that Determining the size type of the object to be loaded according to the size information of the object to be loaded includes: Get the size information of the items to be loaded; Obtaining the similarity of size information between the objects to be loaded; Classifying the objects to be loaded according to the similarity of the size information; Get the size type corresponding to the item to be loaded.

4. The method according to claim 1, It is characterized in that The optimizing the initial loading plan by generating a model through a preset column to obtain a two-dimensional loading plan comprises: Obtain the optimization result of the restricted main problem corresponding to the initial loading plan Obtaining the optimal solution of the dual problem of the master problem in the preset column generation model according to the optimization result of the restricted master problem; According to the optimal solution, generating a loading alternative plan by presetting the price sub-problem in the column generation model; A two-dimensional loading plan is obtained according to the loading alternative plan.

5. The method according to claim 1, It is characterized in that After optimizing the initial loading plan by using the preset column generation model to obtain the two-dimensional loading plan, the method further includes: The two-dimensional loading scheme is optimized by using a preset main heuristic model to obtain an optimized loading scheme; Feedback on the loading optimization plan.

6. A two-dimensional loading optimization device, It is characterized in that The device comprises: A data acquisition module is used to obtain the size information of the items to be loaded; An initial plan acquisition module, used to acquire an initial loading plan according to the size information and preset freight vehicle information; A model building module is used to construct a restricted main problem of a preset column generation model with the goal of minimizing the number of vehicles used; construct a dual problem of the main problem of the preset column generation model according to the restricted main problem; construct constraint conditions of the price sub-problem according to preset two-dimensional loading basic constraints, and the preset two-dimensional loading basic constraints include placement constraints, direction position relationship constraints, vehicle size constraints, coordinate position relationship constraints, loading quantity constraints and maximum loading area constraints; construct an objective function of the price sub-problem with the constraint that violates the dual problem of the main problem to the greatest extent; construct a price sub-problem according to preset model parameters, preset decision variables, the constraint conditions and the objective function; construct a preset column generation model according to the restricted main problem, the dual problem of the main problem and the price sub-problem; A scheme optimization module is used to optimize the initial loading scheme through a preset column generation model to obtain a two-dimensional loading scheme. The preset column generation model is used to generate a loading alternative scheme based on the initial loading scheme, and to find a two-dimensional loading scheme that uses the least number of vehicles from the loading alternative schemes.

7. The device according to claim 6, It is characterized in that The initial plan acquisition module is specifically used to: determine the size type of the items to be loaded according to the size information of the items to be loaded; construct loading plans corresponding to the items to be loaded of each size type according to the preset freight vehicle information; and determine the initial loading plan according to the loading plans corresponding to the items to be loaded of each size type.

8. The device according to claim 7, It is characterized in that The initial solution acquisition module is specifically used to: acquire size information of the objects to be loaded; acquire similarity of size information between the objects to be loaded; classify the objects to be loaded according to the similarity of the size information; and acquire size types corresponding to the objects to be loaded.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images