Logistics scheduling decision-making method, device and equipment

Abstract

The invention provides a logistics scheduling decision-making method, device and equipment, and relates to the technical field of logistics scheduling, and the method comprises the steps: receiving a logistics scheduling decision-making request of a user; based on the logistics scheduling decision request, obtaining current decision associated data of the user; obtaining historical decision association data, and determining target difference data and a target data difference quantity based on the current decision association data and the historical decision association data; if the target data difference quantity meets a preset difference quantity condition, updating a decision model corresponding to the historical decision association data based on the target difference data to obtain an updated decision model; wherein the decision model is constructed based on a mathematical programming algorithm framework; and solving the decision variable of the updated decision model through a prepared optimization mode to obtain a decision result. According to the scheme, modeling is carried out without using all current decision-making associated data, the solving process of the updated decision-making model is simplified, and the decision-making efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics scheduling, in particular to a logistics scheduling decision method, device and equipment. BACKGROUND

[0002] Operations Research (OR) is a cross-disciplinary subject that uses advanced mathematical modeling, statistical analysis, and algorithm design to quantitatively analyze and optimize complex systems. Its core goal is to find the optimal or near-optimal solution under limited resource constraints by establishing a mathematical model, and it is widely used in fields such as logistics scheduling, production planning, financial investment, and energy management.

[0003] In related technologies, the operations research algorithm is usually divided into six serial steps: data reading, data preprocessing, solver tool modeling, solver solving, result post-processing, and result output. When calculating different examples, the full amount of the above six steps needs to be run each time, and the final optimal decision result is obtained.

[0004] For the field of logistics scheduling, especially in the scenario where users need to frequently fine-tune and calculate, the decision efficiency of the above method is low. SUMMARY

[0005] In view of the above problems, the present application aims to provide a logistics scheduling decision method, device and equipment to solve the problem of low decision efficiency of related technologies in the scenario where users need to frequently fine-tune and calculate.

[0006] In a first aspect, the present application provides a logistics scheduling decision method, comprising:

[0007] receiving a logistics scheduling decision request of a user; based on the logistics scheduling decision request, obtaining current decision associated data of the user;

[0008] obtaining historical decision associated data, and based on the current decision associated data and the historical decision associated data, determining target difference data and a target data difference amount;

[0009] If the target data difference amount meets a preset difference amount condition, updating a decision model corresponding to the historical decision associated data based on the target difference data to obtain an updated decision model; wherein the decision model is constructed based on a mathematical programming algorithm framework;

[0010] solving the decision variables of the updated decision model by a prepared optimization method to obtain a decision result; the decision result is used for logistics scheduling.

[0011] In a possible implementation, the target difference data includes basic information and data update attributes; and the updating of the decision model corresponding to the historical decision-related data based on the target difference data to obtain an updated decision model includes:

[0012] updating the decision model corresponding to the historical decision-related data based on the basic information and the data update attributes of the target difference data to obtain an updated decision model.

[0013] In a possible implementation, the basic information includes a name and a value, and the updating of the decision model corresponding to the historical decision-related data based on the basic information and the data update attributes of the target difference data to obtain an updated decision model includes:

[0014] if it is determined that the data update attribute of the target difference data is change, locating a first model parameter in the decision model corresponding to the historical decision-related data based on the name of the target difference data;

[0015] updating a value of the first model parameter based on the value of the target difference data to obtain an updated decision model;

[0016] wherein the first model parameter belongs to a constraint condition.

[0017] In a possible implementation, the updating of the decision model corresponding to the historical decision-related data based on the basic information and the data update attributes of the target difference data to obtain an updated decision model includes:

[0018] if it is determined that the data update attribute of the target difference data is addition, adding a second model parameter based on the basic information of the target difference data to the decision model corresponding to the historical decision-related data to obtain an updated decision model;

[0019] wherein the second model parameter belongs to any one of a variable and a constraint condition.

[0020] In a possible implementation, the basic information includes a name, and the updating of the decision model corresponding to the historical decision-related data based on the basic information and the data update attributes of the target difference data to obtain an updated decision model includes:

[0021] if it is determined that the data update attribute of the target difference data is deletion, locating a third model parameter in the decision model corresponding to the historical decision-related data based on the name of the target difference data;

[0022] deleting the third model parameter in the decision model to obtain an updated decision model;

[0023] The third model parameter belongs to any one of the following: a variable, a constraint condition.

[0024] In a possible implementation, the historical decision correlation data is acquired, and the target difference data and the target data difference amount are determined based on the current decision correlation data and the historical decision correlation data, including:

[0025] The historical decision correlation data is acquired.

[0026] The current decision correlation data is compared with any historical decision correlation data to obtain difference data, and the data difference amount is calculated based on the difference data.

[0027] The smallest data difference amount is selected as the target data difference amount, and the difference data corresponding to the target data difference amount is selected as the target difference data.

[0028] In a possible implementation, before the decision model corresponding to the historical decision correlation data is updated based on the target difference data, the method further includes:

[0029] The target difference data is preprocessed.

[0030] In a second aspect, the application provides a logistics scheduling decision device, the device includes:

[0031] A receiving unit is configured to receive a logistics scheduling decision request of a user.

[0032] An acquisition unit is configured to acquire current decision correlation data of the user based on the logistics scheduling decision request.

[0033] The acquisition unit is further configured to acquire historical decision correlation data.

[0034] A determination unit is configured to determine target difference data and a target data difference amount based on the current decision correlation data and the historical decision correlation data.

[0035] A model updating unit is configured to update a decision model corresponding to the historical decision correlation data based on the target difference data to obtain an updated decision model, if the target data difference amount meets a preset difference amount condition, wherein the decision model is constructed based on a mathematical programming algorithm framework.

[0036] A decision unit is configured to solve a decision variable of the updated decision model by a prepared optimization method to obtain a decision result, wherein the decision result is used for logistics scheduling.

[0037] In a third aspect, an electronic device is provided, including a processor and a memory connected with the processor in communication.

[0038] The memory stores computer-executable instructions;

[0039] The processor executes the computer-executable instructions stored in the memory to implement the method in any possible implementation of the first aspect.

[0040] In a fourth aspect, the embodiments of the present application further provide a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the method in any possible implementation of the first aspect is implemented.

[0041] According to a fifth aspect of the present application, a computer program product is provided, which comprises a computer program. When the computer program is executed by a processor, the method in any possible implementation of the first aspect is implemented.

[0042] The logistics scheduling decision method, device and equipment provided by the present application, the method comprises: receiving a logistics scheduling decision request of a user; based on the logistics scheduling decision request, obtaining current decision associated data of the user; obtaining historical decision associated data, and based on the current decision associated data and the historical decision associated data, determining target difference data and a target data difference amount; if the target data difference amount meets a preset difference amount condition, updating a decision model corresponding to the historical decision associated data based on the target difference data to obtain an updated decision model; wherein the decision model is constructed based on a mathematical programming algorithm framework; the decision variable of the updated decision model is solved by a prepared optimization method to obtain a decision result; the decision result is used for logistics scheduling. The present application determines the target difference data and the target data difference amount by comparing the current decision associated data and the historical decision associated data. In the case that the target data difference amount meets the preset difference amount condition, the historical decision associated data corresponding to the decision model can be directly updated based on the target difference data to obtain the updated decision model. Then the updated decision model is used to determine the decision result. Especially in the scene that the user needs to frequently fine-tune the calculation, the present application does not need to use all the current decision associated data for modeling, simplifies the solving process of the updated decision model, and improves the decision efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0044] Figure 1 A flowchart of a logistics scheduling decision method provided by the first embodiment of the present application is shown in the figure;

[0045] Figure 2 A flowchart of another logistics scheduling decision method provided by the second embodiment of the present application is shown in the figure;

[0046] Figure 3 This is a flowchart illustrating another logistics scheduling decision-making method provided in Embodiment 2 of this application;

[0047] Figure 4 This is a schematic diagram of the structure of a logistics scheduling decision-making device provided in Embodiment 3 of this application;

[0048] Figure 5 This is a hardware structure diagram of an electronic device provided in Embodiment 4 of this application.

[0049] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments.

[0050] Explanation of reference numerals in the attached figures:

[0051] 400 - Logistics scheduling decision-making device; 401 - Receiving unit; 402 - Acquisition unit; 403 - Determination unit; 404 - Model update unit; 405 - Decision unit;

[0052] 501 - Processor; 402 - Memory; 503 - Communication interface; 504 - Communication bus. Detailed Implementation

[0053] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0054] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner. In the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more.

[0055] Operations Research (OR) is an interdisciplinary field that uses mathematical modeling, statistical analysis, and algorithm design to perform quantitative analysis and decision optimization on complex systems. Its core objective is to find optimal or near-optimal solutions under limited resource constraints by establishing mathematical models. It is widely applied in fields such as logistics scheduling, production planning, financial investment, and energy management.

[0056] Mathematical modeling is a fundamental tool in operations research and optimization. It involves abstracting key elements of real-world problems to construct mathematical expressions consisting of decision variables, objective functions, and constraints. Common model types include:

[0057] Linear programming: Models where both the objective function and constraints are linear, applicable to problems such as resource allocation, and usually solved using the simplex method or interior point method;

[0058] Integer programming: Models that require decision variables to be integers, used in discrete decision-making scenarios (such as path planning), and need to be combined with techniques such as branch and bound;

[0059] Nonlinear programming deals with problems where the objective or constraints are nonlinear, relying on gradient descent, heuristic algorithms, etc.

[0060] Dynamic programming: solves multi-stage decision problems and achieves recursive optimization through the Bellman equation;

[0061] Combinatorial optimization: For discrete structures (such as graph theory and scheduling problems), metaheuristic algorithms (such as genetic algorithms and simulated annealing) are often used.

[0062] Solvers are core tools in operations research for automating mathematical model computation. They transform mathematical models into executable computational processes using built-in algorithms, outputting optimal or feasible solutions. Depending on the problem complexity and model type, various solver categories offer multi-language development interfaces. Different programming languages ​​can utilize the proprietary tools of different solvers to complete the mathematical modeling and solving process, allowing developers to flexibly invoke core algorithms according to project requirements.

[0063] Model Files: In operations research and optimization, model files serve as a crucial bridge between mathematical problems and solvers, defining the problem's variables (decision variables, auxiliary variables), objective function, and constraints. The decision model in this solution is the model file. Common standardized model file formats include:

[0064] LP format (Linear Programming): Describes linear programming problems in readable text form, suitable for simple linear models.

[0065] MPS format (Mathematical Programming System): An industry standard proposed by International Business Machines Corporation (IBM), which supports the storage of large-scale optimization problems and is suitable for commercial solvers (such as CPLEX and Gurobi).

[0066] In related technologies, operations research optimization algorithms typically consist of six sequential steps: data reading, data preprocessing, solver tool modeling, solver solving, result post-processing, and result output. Each computational example requires a full run of these six steps to obtain the desired optimal decision result. Specifically, data reading refers to reading the decision-related data input by the user. Data preprocessing involves preprocessing the read decision-related data, including data logic verification, outlier handling, blank value handling, and data logic imputation. Solver tool modeling involves creating a decision model. Solver solving involves using the solver to find the optimal solution for the decision variables in the updated decision model, thus obtaining the decision result. Result post-processing involves processing the decision result into a report required by the client. Result output involves displaying the decision result.

[0067] In the field of logistics scheduling, especially in scenarios where users need to make frequent fine-tuning calculations, the decision-making efficiency of the above methods is relatively low.

[0068] To address the aforementioned technical problems, this application provides a logistics scheduling decision-making method, apparatus, and device. This solution determines the target difference data and the target data difference amount by comparing current decision-related data with historical decision-related data. When the target data difference amount meets a preset difference amount condition, the decision model corresponding to the historical decision-related data can be directly updated using the target difference data to obtain the updated decision model. The updated decision model is then used to determine the decision result. Especially in scenarios where users need to frequently perform fine-tuning calculations, this solution eliminates the need to use all current decision-related data for modeling, simplifying the solution process for the updated decision model and improving decision-making efficiency.

[0069] Figure 1 This is a flowchart illustrating a logistics scheduling decision-making method provided in Embodiment 1 of this application. This embodiment can be applied to logistics scheduling decision-making scenarios. Furthermore, the decision-making method of this solution can also be applied to decision-making scenarios in fields such as production planning, financial investment, and energy management.

[0070] Logistics scheduling decision-making methods can be executed by a logistics scheduling decision-making device, which can be implemented in software and / or hardware and specifically configured in electronic equipment. For example... Figure 1As shown, the method includes:

[0071] Step 101: Receive the user's logistics scheduling decision request; based on the logistics scheduling decision request, obtain the user's current decision-related data.

[0072] Taking the nationwide transportation and scheduling of mineral resources as an example, numerous supply nodes (mines), demand nodes (factories), and transit nodes (warehouses) exist across the country. Mineral resources can be transported and scheduled uniformly through various modes of transport, including air, water, rail, and land, forming an extremely complex nationwide transportation network. This network connects supply nodes => demand nodes, supply nodes => transit nodes, and transit nodes => demand nodes. Each route has attributes such as transport volume constraints and transportation costs, and each node also has inventory limits. Furthermore, based on factors such as location and province, nodes can be divided into different sets, each set having unified inventory constraints, and transport volume constraints also exist between sets. Therefore, the current decision-making data must at least include the aforementioned transportation network information, the transport volume constraints and transportation costs of each route, as well as the inventory constraints of each set and the transport volume constraints between sets.

[0073] In implementation, a decision model can be created based on the initially input decision-related data, leading to the decision result. This decision-making process requires six complete steps: data reading, data preprocessing, solver tool modeling, solver solving, result post-processing, and result output. The decision model obtained in this process can be used as the root model for subsequent iterations. The decision-related data can be saved, allowing users to modify the data based on this initial data to obtain the current decision-related data. This approach saves users time inputting data.

[0074] In practice, the system can save the user's input of the previous decision-related data, allowing the user to modify the data based on the previous decision-related data to obtain the current decision-related data. This method saves users time when inputting data.

[0075] Furthermore, the data associated with the previous decision can be used as historical decision-related data, and the modifications made by the user based on the previous decision-related data can be used as target difference data. This method eliminates the need to compare the current decision-related data with historical decision-related data, saving decision-making time in this solution.

[0076] Step 102: Obtain historical decision-related data, and based on the current decision-related data and historical decision-related data, determine the target difference data and the target data difference amount.

[0077] In implementation, current decision-related data and historical decision-related data can be compared to obtain the difference data between the current decision-related data and historical decision-related data, and this difference data can be used as the target difference data. The ratio of the target difference data to the historical decision-related data can be used as the target data difference quantity.

[0078] Step 103: If the target data difference meets the preset difference condition, the decision model corresponding to the historical decision-related data is updated based on the target difference data to obtain the updated decision model; wherein, the decision model is constructed based on the mathematical programming algorithm framework.

[0079] The preset difference condition can be that the difference in the target data is less than a preset threshold. For example, this preset threshold can be set to a value such as 5% or 10% as needed.

[0080] The smaller the difference in the target data, the higher the similarity between the current decision-related data and the historical decision-related data.

[0081] In practice, target difference data can represent the modifications made to the current decision-related data relative to historical decision-related data. First, the names of the parameters contained in the target difference data can be used to locate the model parameters in the decision model that need to be modified, thereby updating the decision model and obtaining the updated model.

[0082] The mathematical programming algorithm framework can include mixed integer models, integer models, linear models, nonlinear models, and so on.

[0083] Step 104: Solve the decision variables of the updated decision model using the pre-defined optimization method to obtain the decision results; the decision results are used for logistics scheduling.

[0084] The "prepared optimization method" refers to a pre-prepared optimization method. This scheme does not restrict the optimization method. For example, the optimization method can be a solver or a heuristic method. The solver can utilize mature commercial solvers, such as the Cardinal Optimizer (COPT).

[0085] The logistics scheduling decision-making method provided in the above embodiments determines the target difference data and the target data difference amount by comparing the current decision-related data and the historical decision-related data. When the target data difference amount meets a preset difference amount condition, the decision model corresponding to the historical decision-related data can be directly updated using the target difference data to obtain the updated decision model. The updated decision model is then used to determine the decision result. Especially in scenarios where users need to frequently perform fine-tuning calculations, this solution eliminates the need to use all current decision-related data for modeling, simplifying the solution process for the updated decision model and improving decision-making efficiency.

[0086] Figure 2 This is a flowchart illustrating another logistics scheduling decision-making method provided in Embodiment 2 of this application. This embodiment is... Figure 1 Based on the illustrated embodiment, improvements have been made to the logistics scheduling decision-making method.

[0087] like Figure 2 As shown, a logistics scheduling decision-making method may include the following steps:

[0088] Step 201: Receive the user's logistics scheduling decision request; based on the logistics scheduling decision request, obtain the user's current decision-related data.

[0089] In practice, the principles and implementation methods of step 201 are similar to those of step 101, and will not be repeated here.

[0090] Step 202: Obtain multiple historical decision-related data; compare the current decision-related data with any historical decision-related data to obtain the difference data; and calculate the data difference quantity based on the difference data.

[0091] In practice, decision-related data input by the user each time can be saved, and multiple historical decision-related data can be selected for use. For example, the decision-related data from the five most recent user inputs can be selected as five historical decision-related data sets for use.

[0092] In implementation, the data type of the discrepancy data can be determined first. Different data types (such as supply, demand, inventory, etc.) use different data tables, so the location can be directly determined. Then, the number of discrepancy rows can be determined. Since each data table has only one row of data per node, modifications to that row can be directly retrieved. The ratio between the number of discrepancy rows and the total number of data rows can be used as the data discrepancy quantity.

[0093] Step 203: Select the smallest data difference as the target data difference; and use the difference data corresponding to the target data difference as the target difference data.

[0094] In practice, the data with the smallest difference from the current decision-related data can be selected from multiple historical decision-related data, thereby reducing the amount of data with target differences. This allows for model creation with less data, further improving decision-making efficiency.

[0095] Step 204: If the target data difference meets the preset difference condition, then perform data preprocessing on the target difference data.

[0096] Taking the nationwide transportation and scheduling of mineral resources as an example, firstly, the data type of the target variance data can be determined. This data type can include supply, demand, transportation, inventory data, and aggregated data from various nodes. Different data types have corresponding preprocessing logic. Based on the data type of the target variance data, the corresponding preprocessing logic can be determined, and then the appropriate preprocessing operations can be performed.

[0097] For example, if the data type of the target difference data is supply data, then data validation processing can be performed on the target difference data to handle and restore anomalies such as negative initial inventory values ​​or the minimum value of supply constraints being greater than the maximum value.

[0098] Data logic is used to populate the target difference data, just as the customer agrees that when the transportation limit is empty, it can be considered unlimited;

[0099] Supply data processing involves processing the shared inventory of the supply node set, as well as the inventory capacity, supply capacity, and transportation capacity of each node. This results in a complete combination of node limits, inventory upper and lower limits, supply capacity upper and lower limits, and transportation capacity upper and lower limits.

[0100] Preprocessing the target difference data can improve its accuracy, thereby improving decision-making accuracy. Furthermore, this solution does not require preprocessing all current decision-related data; it only preprocesses the target difference data, thus improving decision-making efficiency.

[0101] Step 205: Based on the basic information of the target difference data and the data update attributes, update the decision model corresponding to the historical decision association data to obtain the updated decision model; wherein, the decision model is constructed based on the mathematical programming algorithm framework.

[0102] The basic information may include the name and value of the target difference data.

[0103] The data update attribute is used to characterize the type of change that has occurred in the current decision-related data compared to the historical decision-related data. The data update attribute can include addition, deletion, and modification.

[0104] For example, based on the name of the target difference data, the model parameters that need to be updated in the decision model can be located. Then, based on the value of the target difference data and the data update attributes, the model parameters can be updated accordingly to obtain the updated decision model.

[0105] The above method can conveniently enable the creation of updated decision-making models.

[0106] In one feasible approach, the basic information includes name and value. Based on the basic information of the target difference data and data update attributes, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model, including:

[0107] If the data update attribute of the target difference data is determined to be "change", then based on the name of the target difference data, the first model parameter in the decision model corresponding to the historical decision association data is located; the value of the first model parameter is updated based on the value of the target difference data to obtain the updated decision model; wherein, the first model parameter is a constraint condition.

[0108] For example, the target variance data could be named the supply ceiling of node A in the supply data. The data update attribute could be set to "Change." Then, based on the value of the target variance data, the supply ceiling of node A in the decision model could be modified.

[0109] In one feasible approach, based on the basic information of the target difference data and the data update attributes, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model, including:

[0110] If the data update attribute of the target difference data is determined to be new, then based on the basic information of the target difference data, a second model parameter is added to the decision model corresponding to the historical decision association data to obtain the updated decision model; wherein, the second model parameter belongs to any of the following: variable, constraint condition.

[0111] For example, the target difference data could be named the supply ceiling of node B in the supply data, and the data update attribute could be set to "new." Then, based on the basic information of the target difference data, a supply ceiling constraint for node B could be added to the decision model.

[0112] In one possible implementation, the basic information includes the name, basic information based on the target difference data, and data update attributes. The decision model corresponding to the historical decision-related data is then updated to obtain the updated decision model, which includes:

[0113] If the data update attribute of the target difference data is determined to be deletion, then based on the name of the target difference data, locate the third model parameter in the decision model corresponding to the historical decision association data; delete the third model parameter in the decision model to obtain the updated decision model; wherein, the third model parameter belongs to any of the following: variable, constraint condition.

[0114] For example, the target difference data could be named the supply ceiling constraint at node C in the supply data. If the data update attribute is set to delete, then the supply ceiling constraint at node C in the decision model can be deleted.

[0115] The above method can conveniently enable the creation of updated decision-making models.

[0116] Step 206: Solve the decision variables of the updated decision model using the pre-prepared optimization method to obtain the decision results; the decision results are used for logistics scheduling.

[0117] In practice, the principle and implementation method of step 206 are similar to those of step 104, and will not be repeated here.

[0118] refer to Figure 3 The relevant technical decision-making methods include the following steps: full data reading, full data preprocessing, full modeling using solver tools, solver call, result post-processing, and data writing to database / file system.

[0119] refer to Figure 3 This solution can run the above steps only once during the initial run. When the user performs fine-tuning calculations on the decision model, it is not necessary to run the above steps again. By comparing the current decision-related data with the historical decision-related data, the target difference data is obtained. Then, difference data processing is performed on the target difference data, i.e., data preprocessing. Then, the decision model corresponding to the historical decision-related data is updated using the preprocessed target difference data, resulting in the updated decision model (i.e.,...). Figure 3 (The model file in the file). Next, the updated decision model can be used to perform solver calls, result post-processing, and data writing to the database / file system.

[0120] refer to Figure 3 This invention primarily addresses the resource and time waste caused by repeated similar runs of optimization algorithms, and mainly consists of the following steps:

[0121] 1. Obtain the initial model file: In order to ensure that the subsequent iteration process has a root model, the algorithm can be run in its entirety with any version as the initial model in the same set of similar input data (i.e. decision association data). This run needs to go through six steps: data reading, data preprocessing, solver tool modeling, solver solving, result postprocessing, and result output.

[0122] During the modeling process using solver tools, it's important to ensure standardized naming when creating variables, constraints, and objective functions. The names of different variables and constraints must be unique to facilitate later identification. Another difference is that after modeling with solver tools, the mathematical model needs to be saved offline using the solver tool's functionality. This involves outputting and saving the model file, using formats such as lp and mps.

[0123] 2. Compare the differences between multiple versions of input data, or require users to input only the updated parts. Because in the scenario of this invention, the customer performs multiple fine-tuning and optimization algorithm calculations, and the amount of data adjusted in each calculation is much smaller than the total amount of data. Therefore, the difference can be separated using a data difference comparison module, or the customer can directly input the adjusted parts each time.

[0124] 3. Differential Data Processing: The processing logic in the input data needs to be reproduced on the differential data. However, due to the significant reduction in data volume, the time required for this step will be greatly optimized. Finally, the output will contain differential data for three different differential categories: variable / constraint deletion, constraint modification (variables that cannot be modified can be achieved by deleting and then adding them), and variable / constraint addition.

[0125] 4. Root Model Reading: This step involves reading the model file generated in the first step. Each solver tool provides a corresponding interface. After reading, the complete decision model can be obtained, which includes all variables, constraints, objective functions, and their names.

[0126] 5. Difference Model Construction: Based on the difference data processed in step 3, the decision model is modified to address the differences. This process can utilize data, variables, and constraint names for precise localization, and is handled separately according to different difference types (i.e., data update attributes):

[0127] For variable / constraint deletion: Variable deletion can force the original variable to be assigned a value of 0 to avoid affecting other constraints containing this variable; constraint deletion can be done directly in the model.

[0128] For the changed parts: directly modify the original constraints and update the constant terms in the constraints;

[0129] For newly added variables / constraints: new variables and constraints can be built directly on the basis of the basic model, and the newly added variables can be incorporated into the objective function.

[0130] 6. Solving the difference model: After completing the construction of the difference model (i.e., the updated decision model), the optimal solution can be obtained by solving the model using the solver.

[0131] 7. Post-processing of results: Generate and output result reports according to user requirements.

[0132] This invention can bring the following beneficial effects:

[0133] By using model files to save the root model, the final results of many preceding tasks can be preserved and avoided in subsequent runs.

[0134] The model file is permanently stored offline, so there is a high degree of reusability. For example, if the difference between any evaluation run and the base run is less than the full dataset, the same model file can be used for processing and calculation.

[0135] By leveraging the differences in data to build different models, the significant difference in data volume between the different parts and the full parts can greatly reduce the time cost of each step.

[0136] Since the differences between each run are small, the data processing, variable construction, and constraint construction performed in the first run can be omitted in subsequent runs. The model file, which can completely save the results of all previous steps, becomes the basis for realizing difference updates.

[0137] Since the base model file can be reused multiple times, the more similar data examples there are, the greater the time cost benefit that this method can generate.

[0138] For the difference portion, the greater the difference between the amount of difference data and the amount of full data, the more redundant calculations this method can save, and therefore the higher the efficiency improvement achieved by this method.

[0139] For scenarios where multiple input data are similar and only a very small portion varies, running the algorithm in its entirety every time is not the most economical or practical approach. The problem that needs to be solved is how to perform algorithmic calculations based solely on the differences between models, given a basic model in place.

[0140] For scenarios where multiple input data are similar and only a very small portion varies, running the algorithm in its entirety every time is not the most economical and practical approach. This solution, based on a root model, allows algorithm calculation solely based on the differences between models, thus improving decision-making efficiency.

[0141] Figure 4 This is a schematic diagram of a logistics scheduling decision-making device provided in Embodiment 3 of this application. The device can be in the form of software and / or hardware. (See also...) Figure 4 As shown, a logistics scheduling decision-making device 400 includes: a receiving unit 401, an acquisition unit 402, a determination unit 403, a model update unit 404, and a decision-making unit 405.

[0142] The receiving unit 401 is used to receive the user's logistics scheduling decision request;

[0143] The acquisition unit 402 is used to acquire the user's current decision-related data based on the logistics scheduling decision request;

[0144] The acquisition unit 402 is also used to acquire historical decision-related data;

[0145] The determining unit 403 is used to determine the target difference data and the target data difference amount based on the current decision-related data and the historical decision-related data;

[0146] The model update unit 404 is used to update the decision model corresponding to the historical decision-related data based on the target difference data if the target data difference meets the preset difference condition, so as to obtain the updated decision model; wherein, the decision model is constructed based on the mathematical programming algorithm framework.

[0147] The decision unit 405 is used to solve the decision variables of the updated decision model through a pre-defined optimization method to obtain the decision results; the decision results are used for logistics scheduling.

[0148] The model update unit 404 is specifically used for target difference data, including basic information and data update attributes; based on the basic information and data update attributes of the target difference data, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model.

[0149] The model update unit 404 is specifically used for basic information including name and value. If the data update attribute of the target difference data is determined to be change, the first model parameter in the decision model corresponding to the historical decision-related data is located based on the name of the target difference data.

[0150] The values ​​of the first model parameters are updated based on the values ​​of the target difference data to obtain the updated decision model;

[0151] Among them, the first model parameter is a constraint condition.

[0152] The model update unit 404 is specifically used to add a second model parameter to the decision model corresponding to the historical decision association data based on the basic information of the target difference data if the data update attribute of the target difference data is determined to be new, so as to obtain the updated decision model.

[0153] The second model parameter can be any of the following: variable or constraint.

[0154] The model update unit 404 is specifically used for basic information including name. If the data update attribute of the target difference data is determined to be deletion, the third model parameter in the decision model corresponding to the historical decision association data is located based on the name of the target difference data.

[0155] The third model parameter in the decision model is removed to obtain the updated decision model;

[0156] The third model parameter can be any of the following: variable or constraint.

[0157] The acquisition unit 402 is specifically used to acquire multiple historical decision-related data.

[0158] The determining unit 403 is specifically used to compare the current decision-related data with any historical decision-related data to obtain the difference data; and to calculate the data difference quantity based on the difference data.

[0159] Select the smallest data difference as the target data difference; and use the difference data corresponding to the target data difference as the target difference data.

[0160] The determination unit 403 is also used to perform data preprocessing on the target difference data.

[0161] The logistics scheduling decision-making device provided in this application embodiment has the same implementation principle and technical effect as the aforementioned logistics scheduling decision-making method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned logistics scheduling decision-making method embodiment.

[0162] Figure 5 This is a hardware structure diagram of an electronic device provided in Embodiment 4 of this application. This embodiment provides an electronic device including: at least one processor 501, and a memory 502 communicatively connected to the at least one processor 501; the memory 502 stores computer-executable instructions; the processor 501 executes the computer-executable instructions stored in the memory 502 to implement the logistics scheduling decision-making method described in any of the preceding embodiments.

[0163] Figure 5The illustrated electronic device also includes a communication interface 503 and a communication bus 504. The processor 501, memory 502, and communication interface 503 are interconnected via the communication bus 504. The communication bus 504 can be divided into an address bus, a data bus, a control bus, etc., for ease of representation. Figure 5 The communication bus 504 is represented by a single thick line, but this does not mean that there is only one communication bus 504 or only one type of communication bus 504. The processor 501 can also be called a controller; there is no restriction on the name.

[0164] In this embodiment, the memory 502 stores instructions executable by at least one processor 501. By executing the instructions stored in the memory 502, the at least one processor 501 can execute the logistics scheduling decision-making method discussed above. The processor 501 can implement... Figure 5 The functions of each module in the device shown.

[0165] The processor 501 is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory 502 and calling data stored in memory 502, the processor can perform various functions and process data, thereby monitoring the device as a whole.

[0166] In one possible design, processor 501 may include one or more processing units. Processor 501 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into processor 501. In some embodiments, processor 501 and memory 502 may be implemented on the same chip or on separate chips.

[0167] Processor 501 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the logistics scheduling decision-making method disclosed in the embodiments of this application can be directly manifested as execution by a hardware processor, or as execution by a combination of hardware and software modules within the processor.

[0168] Memory 502, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 502 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic storage, magnetic disk, optical disk, etc. Memory 502 can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. In the embodiments of this application, memory 502 can also be a circuit or any other device capable of implementing storage functions for storing program instructions and / or data.

[0169] By designing and programming the processor 501, the code corresponding to the logistics scheduling decision-making method described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the code during runtime. Figure 1 or Figure 2 The steps of the logistics scheduling decision-making method in the illustrated embodiment are as follows. How to design and program the processor 501 is a technique well-known to those skilled in the art and will not be described further here.

[0170] This application also provides a computer-readable storage medium storing computer-executable instructions. When executed by a processor, these instructions are used to implement the logistics scheduling decision-making method described in any of the preceding embodiments; therefore, they will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the embodiments of the computer storage medium involved in this invention, please refer to the description of the method embodiments of this invention.

[0171] In some possible implementations, various aspects of the logistics scheduling decision-making method provided in this application can also be implemented in the form of a program product, which includes program code. When the program product is run on a device, the program code is used to cause the control device to perform the steps in the logistics scheduling decision-making method according to the various exemplary embodiments of this application described above.

[0172] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0173] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0174] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0175] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0176] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A logistics scheduling decision-making method, characterized in that, The method includes: Receive logistics scheduling decision requests from users; based on the logistics scheduling decision requests, obtain the user's current decision-related data; Acquire historical decision-related data, and based on current decision-related data and historical decision-related data, determine the target difference data and the target data difference amount; If the target data difference meets the preset difference condition, the decision model corresponding to the historical decision-related data is updated based on the target difference data to obtain the updated decision model; wherein, the decision model is constructed based on the mathematical programming algorithm framework; The decision variables of the updated decision model are solved using a pre-defined optimization method to obtain the decision results; these decision results are then used for logistics scheduling.

2. The method according to claim 1, characterized in that, The target difference data includes basic information and data update attributes; then, updating the decision model corresponding to the historical decision-related data based on the target difference data to obtain the updated decision model includes: Based on the basic information of the target difference data and the data update attributes, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model.

3. The method according to claim 2, characterized in that, The basic information includes name and value. Based on the basic information of the target difference data and the data update attributes, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model, including: If the data update attribute of the target difference data is determined to be "change", then based on the name of the target difference data, locate the first model parameter in the decision model corresponding to the historical decision association data; The values ​​of the first model parameters are updated based on the values ​​of the target difference data to obtain the updated decision model; The first model parameter is a constraint condition.

4. The method according to claim 2, characterized in that, The decision model corresponding to historical decision-related data is updated based on the basic information and data update attributes of the target difference data to obtain the updated decision model, including: If the data update attribute of the target difference data is determined to be new, then based on the basic information of the target difference data, a second model parameter is added to the decision model corresponding to the historical decision association data to obtain the updated decision model. The second model parameter can be any of the following: variable or constraint.

5. The method according to claim 2, characterized in that, The basic information includes a name. Based on the basic information of the target difference data and data update attributes, the decision model corresponding to the historical decision-related data is updated to obtain the updated decision model, including: If the data update attribute of the target difference data is determined to be deletion, then based on the name of the target difference data, the third model parameter in the decision model corresponding to the historical decision association data is located. The third model parameter in the decision model is removed to obtain the updated decision model; The third model parameter can be any of the following: variable or constraint.

6. The method according to claim 1, characterized in that, The step of acquiring historical decision-related data and determining the target difference data and the target data difference amount based on the current decision-related data and historical decision-related data includes: Obtain correlation data from multiple historical decisions; The current decision-related data is compared with any historical decision-related data to obtain the difference data; and the data difference quantity is calculated based on the difference data. Select the smallest data difference as the target data difference; and use the difference data corresponding to the target data difference as the target difference data.

7. The method according to claim 1, characterized in that, Before updating the decision model corresponding to historical decision-related data based on the target difference data, the following is also included: Perform data preprocessing on the target difference data.

8. A logistics scheduling decision-making device, characterized in that, The device includes: The receiving unit is used to receive users' logistics scheduling decision requests; The acquisition unit is used to acquire the user's current decision-related data based on the logistics scheduling decision request; The acquisition unit is also used to acquire historical decision-related data; The determination unit is used to determine the target difference data and the target data difference amount based on the current decision-related data and historical decision-related data; The model update unit is used to update the decision model corresponding to the historical decision-related data based on the target difference data if the target data difference meets the preset difference condition, so as to obtain the updated decision model; wherein, the decision model is constructed based on the mathematical programming algorithm framework; The decision-making unit is used to solve the decision variables of the updated decision model through a pre-defined optimization method to obtain the decision results; the decision results are used for logistics scheduling.

9. An electronic device, characterized in that, Includes a processor and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the logistics scheduling decision method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the logistics scheduling decision-making method as described in any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the logistics scheduling decision method according to any one of claims 1-7.

Images