A logistics warehousing management method and system based on digital twin

Abstract

The present invention provides a logistics warehousing management method and system based on digital twins, which relates to the field of warehousing logistics technology. The logistics warehousing management method based on digital twins includes the following steps: obtaining logistics data and structural layout data of the warehousing, and constructing a warehousing digital twin model; obtaining the order data of goods in the warehousing, analyzing the importance of the goods order through a projection tracking algorithm, and prioritizing the goods order through a directed graph analysis method; based on the priority sorting result, optimizing the storage location of the goods through the constructed warehousing digital twin model, and determining the optimal storage location of each product; dynamically adjusting the placement of the goods in the warehousing according to the optimal storage location of each product. The present invention can improve the speed and accuracy of goods out of the warehouse by optimizing the storage location and warehousing process of the goods, thereby shortening the waiting time for customers and improving customer satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of warehousing and logistics technology, and in particular, to a logistics warehousing management method and system based on digital twins. Background Art

[0002] With the rapid development of industrialization and informatization, the logistics industry has become a key pillar in the global economy. As an indispensable part of the logistics system, the improvement of warehousing logistics efficiency and management level is crucial for enterprises to enhance their market competitiveness. Especially in today's era of rapid digital evolution, the adoption of intelligent management strategies is the key to improving the efficiency of warehousing operations. In particular, the intelligent warehouse management system based on the digital twin model can effectively predict and optimize the warehouse layout and logistics path through high-precision simulation and real-time data analysis. This technology can not only greatly improve logistics efficiency, but also significantly reduce operating costs, thereby directly improving the overall performance and market competitiveness of the enterprise. In addition, combined with the path planning algorithm, this management method can further optimize the storage and picking path of goods, reduce human resource consumption and time costs, and ensure a more efficient and accurate order processing process. Therefore, intelligent warehouse logistics management has become an important strategic tool for modern enterprises to improve operational efficiency, enhance customer satisfaction and sustainable competitiveness.

[0003] At present, in the existing warehouse management technology, the utilization efficiency of warehouse space is usually not high; this is mainly due to the lack of efficient space optimization technology and methods, resulting in unreasonable storage locations of goods, resulting in long picking paths and insufficient space utilization; this inefficient space management not only increases the time cost of picking and transportation, but also leads to waste of resources, seriously affecting the overall operation efficiency of the warehouse system. In addition, non-optimized product layout may also lead to logistics congestion and increased operational complexity, further increasing logistics costs.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of this, and in response to the problems in the related art, the present invention provides a logistics warehousing management method and system based on digital twins to solve the above-mentioned problem of low utilization efficiency of warehouse space.

[0006] In order to solve the above problems, the specific technical solutions adopted by the present invention are as follows:

[0007] According to one aspect of the present invention, a logistics warehousing management method based on digital twins is provided, and the logistics warehousing management method based on digital twins comprises the following steps:

[0008] S1. Obtain the logistics data and structural layout data of the warehouse and build a digital twin model of the warehouse;

[0009] S2. Obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection tracking algorithm, and prioritize the goods orders through the directed graph analysis method;

[0010] S3. Based on the priority sorting results, the storage location of the goods is optimized through the constructed warehouse digital twin model to determine the optimal storage location for each product;

[0011] S4. Dynamically adjust the placement of goods in the warehouse based on the optimal storage location of each product.

[0012] Preferably, obtaining the order data of goods in the warehouse, analyzing the importance of the goods orders by a projection pursuit algorithm, and prioritizing the goods orders by a directed graph analysis method comprises the following steps:

[0013] S21. Preprocess the order data of goods in the warehouse using the range transformation method and calculate the projection index parameters, where the projection index parameters include the standard deviation value and the density value;

[0014] S22, optimizing the projection index parameters through the projection index optimization algorithm, and calculating the importance of each product order;

[0015] S23. Construct an initial directed graph, and prioritize the product orders using a graph theory algorithm according to the importance of each product order.

[0016] Preferably, using the range transformation method to preprocess the order data of goods in the warehouse and calculate the projection index parameters includes the following steps:

[0017] S211, performing data cleaning on each indicator data in the goods order data in the warehouse;

[0018] S212, using the range transformation method to normalize each indicator data after data cleaning;

[0019] S213, setting an initial projection direction vector for each indicator data based on a heuristic method, and multiplying the initial weight value by the normalized indicator data to obtain an indicator projection value;

[0020] S214. According to the obtained indicator projection value of each indicator data, the standard deviation value and the density value of the indicator projection value are calculated respectively.

[0021] Preferably, according to the obtained indicator projection value of each indicator data, the calculation formulas for respectively calculating the standard deviation value and the density value of the indicator projection value are as follows:

[0022] ;

[0023] ;

[0024] Where V H Represents the standard deviation of the index projection value of the i-th index data in the product order data;

[0025] H i Represents the index projection value of the i-th index data in the product order data;

[0026] Represents the average value of the index projection values ​​of n index data in the product order data;

[0027] G H The density value representing the index projection value of the i-th index data in the product order data;

[0028] K represents the local width parameter;

[0029] d i,j Represents the difference between the i-th index data in the product order data and the j-th index data in the product order data;

[0030] C(Kd i,j ) represents a unit step function, if Kd i,j =0, the value is 1, otherwise the value is 0.

[0031] Preferably, optimizing the projection index parameters by using the projection index optimization algorithm and calculating the importance of each product order includes the following steps:

[0032] S221, randomly initialize the projection direction vector of each indicator data to generate an initial projection direction vector population;

[0033] S222, for each individual in the initial projection direction vector population, calculate its fitness value, where the fitness value is the product of the standard deviation value and the density value;

[0034] S223, performing selection, crossover and mutation operations on the projection direction vector population according to the fitness value to generate a new projection direction vector population;

[0035] S224, repeating steps S222 to S223 until the preset number of iterations is met, selecting the projection direction vector with the highest fitness as the optimal projection direction vector, and calculating the index projection value of each product order in this direction;

[0036] S225. Use the index projection value of each product order as the importance of each product order.

[0037] Preferably, constructing an initial directed graph and prioritizing the order of goods by a graph theory algorithm according to the importance of each order of goods includes the following steps:

[0038] S231, randomly selecting two groups of product orders as nodes, and establishing edges according to the importance between the two groups of product orders to form an initial directed graph;

[0039] S232, inserting new product order nodes into the initial directed graph one by one according to the importance of the product orders;

[0040] S233, analyzing whether there is a dependency relationship between the new product order node and the product order node in the initial directed graph;

[0041] S234. If there is a dependency relationship, the new product order node is connected to the product order node in the initial directed graph to form an edge; if there is no dependency relationship, the new product order node is compared with the product order node in the initial directed graph in terms of importance, and the new product order node is connected to the product order node in the initial directed graph to form an edge according to the comparison result;

[0042] S235. Repeat steps S232-S234 until all nodes are inserted into the directed graph to obtain a complete directed graph of goods orders, and use the topological sorting method in the graph theory algorithm to prioritize the formed directed graph of goods orders.

[0043] Preferably, based on the priority sorting result, the storage location of the goods is optimized by constructing a warehouse digital twin model, and determining the optimal storage location of each product includes the following steps:

[0044] S31. Integrate the priority sorting results into the warehouse digital twin model and simulate the storage locations of different goods;

[0045] S32, establishing a goods location storage model with minimizing the goods picking path as the objective function;

[0046] S33. Solve the goods location storage model through a location solving algorithm to obtain the optimal storage location of the goods.

[0047] Preferably, solving the goods location storage model by a location solving algorithm to obtain the optimal storage location of the goods includes the following steps:

[0048] S331. Randomly generate a storage plan for the location of goods, and simulate the picking path of each product based on the warehouse digital twin model;

[0049] S332, calculating the picking path lengths of all goods in each storage solution, and sorting the goods according to the picking path lengths to obtain a path sorting result;

[0050] S333, calculating the priority loss value between the path sorting result and the priority sorting result;

[0051] S334. If the priority loss value is lower than the preset threshold, the goods storage location corresponding to the path sorting result is used as the optimal storage location for the goods; otherwise, the storage plan is improved using an iterative optimization algorithm to generate a new storage plan, and steps S332-S334 are repeated until the preset stop condition is met.

[0052] Preferably, calculating the priority loss value between the path sorting result and the priority sorting result comprises the following steps:

[0053] S3321. Determine the sorting position of each product according to the obtained path sorting result and priority sorting result;

[0054] S3322. For each product, calculate the difference in its sorting position in the two sorting results to obtain a sorting difference value;

[0055] S3323. Add the sorting difference values ​​of all goods, and use the added result as the priority loss value between the path sorting result and the priority sorting result.

[0056] According to another aspect of the present invention, a logistics warehousing management system based on digital twin is provided, and the logistics warehousing management system based on digital twin includes: a model building module, a priority analysis module, a storage location determination module and a storage location adjustment module, and the model building module, the priority analysis module, the storage location determination module and the storage location adjustment module are connected in sequence;

[0057] Model building module, used to obtain warehouse logistics data and structural layout data and build a warehouse digital twin model;

[0058] The priority analysis module is used to obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection tracking algorithm, and prioritize the goods orders through the directed graph analysis method;

[0059] The storage location determination module is used to optimize the storage location of goods based on the priority sorting results through the constructed warehouse digital twin model to determine the optimal storage location for each product;

[0060] The storage location adjustment module is used to dynamically adjust the placement of goods in the warehouse according to the optimal storage location of each product.

[0061] Compared with the prior art, the present invention provides a logistics warehousing management method and system based on digital twins, which has the following beneficial effects:

[0062] (1) The present invention realizes accurate simulation and prediction of real warehousing by constructing a digital twin model of warehousing, which helps managers to discover potential problems in advance and optimize warehousing processes. Through digital twin technology, the allocation of warehousing resources can be optimized and unnecessary waste can be reduced. By simulating and optimizing the warehousing process, the investment in manpower, material and financial resources can be reduced, thereby reducing operating costs. By optimizing the storage location and warehousing process of goods, the speed and accuracy of goods out of the warehouse can be improved, thereby shortening customer waiting time and improving customer satisfaction.

[0063] (2) The present invention uses the range transformation method to preprocess and normalize the data, which can eliminate the dimensional differences between different indicators and make the comparison between different indicators more meaningful. Through the projection indicator optimization algorithm, the projection direction vector can be automatically adjusted to find the optimal projection direction, so that the projection indicator parameters can more accurately reflect the importance of the goods order. The importance of the goods order is calculated by combining the standard deviation value and the density value, which can comprehensively consider the discrete degree and distribution characteristics of the order data, making the calculation of the importance more comprehensive and reasonable. The initial directed graph is constructed and nodes are inserted and edges are connected based on the importance of the goods order, which can intuitively display the dependency and priority relationship between the goods orders. The topological sorting method in the graph theory algorithm is used to prioritize the directed graph of the goods order, which can ensure the rationality and accuracy of the sorting result.

[0064] (3) The present invention integrates the priority sorting results into the warehouse digital twin model and simulates the storage locations of different goods. It can accurately predict and evaluate the warehouse efficiency under different storage strategies. Taking minimizing the goods picking path as the objective function, it ensures that while meeting the priority requirements, the storage location of the goods can maximize the reduction of picking time and distance, thereby significantly improving the efficiency and response speed of warehouse operations. The optimized storage location of goods reduces unnecessary handling and movement, and reduces warehouse operating costs, including labor costs, time costs, and equipment loss costs. At the same time, by reducing picking time and distance, it can also reduce fines and return costs caused by delayed delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0066] Figure 1 is a flow chart of a logistics warehousing management method based on digital twins according to an embodiment of the present invention;

[0067] Figure 2 is a principle block diagram of a logistics warehousing management system based on digital twins according to an embodiment of the present invention;

[0068] Figure 3 It is a structural schematic diagram of a computer device according to an embodiment of the present invention.

[0069] In the figure:

[0070] 1. Model building module; 2. Priority analysis module; 3. Storage location determination module; 4. Storage location adjustment module. DETAILED DESCRIPTION

[0071] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0072] According to an embodiment of the present invention, a logistics warehousing management method and system based on digital twins are provided.

[0073] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a logistics warehousing management method based on digital twins is provided, and the logistics warehousing management method based on digital twins includes the following steps:

[0074] S1. Obtain the logistics data and structural layout data of the warehouse and build a digital twin model of the warehouse;

[0075] Specifically, obtaining the logistics data and structural layout data of the warehouse and building a warehouse digital twin model include the following steps:

[0076] Collect information related to warehousing logistics, including but not limited to goods inbound and outbound records, inventory, goods turnover speed, order information, etc.; collect physical layout information of storage facilities, such as shelf location, size, aisle width, door location, etc.;

[0077] Clean the collected logistics data to remove duplicate, erroneous or inconsistent information; convert and standardize the structure layout data to facilitate subsequent model building;

[0078] Use 3D modeling software (such as AutoCAD, Revit, etc.) or a dedicated digital twin platform to create a 3D model of the storage facility based on the collected structural layout data; associate the cleaned logistics data with the 3D model to ensure that the model can accurately reflect the real-time status of the warehouse.

[0079] S2. Obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection tracking algorithm, and prioritize the goods orders through the directed graph analysis method;

[0080] As a preferred implementation, obtaining the order data of goods in the warehouse, analyzing the importance of the goods orders by projection tracking algorithm, and prioritizing the goods orders by directed graph analysis method include the following steps:

[0081] S21. Preprocess the order data of goods in the warehouse using the range transformation method and calculate the projection index parameters, where the projection index parameters include the standard deviation value and the density value;

[0082] As a preferred implementation, using the range transformation method to preprocess the order data of goods in the warehouse and calculate the projection index parameters includes the following steps:

[0083] S211, performing data cleaning on each indicator data in the goods order data in the warehouse;

[0084] It should be noted that data cleaning is an important step to ensure data quality and accuracy, especially when processing various indicators of goods order data in warehousing, including missing value processing, outlier detection and processing, etc.

[0085] S212, using the range transformation method to normalize each indicator data after data cleaning;

[0086] It should be noted that data normalization using the range transformation method (also known as minimum-maximum normalization or linear scaling) is a commonly used data preprocessing technology, which can scale the original data to a specified range, usually between 0 and 1; in the logistics warehousing management method based on digital twins, normalization of the cleaned goods order data indicators helps to eliminate the impact of dimensions and magnitudes between different indicators, making each indicator comparable in data analysis;

[0087] Specifically, the steps of normalization using the range transformation method are:

[0088] For each indicator data after data cleaning, first find the maximum and minimum values ​​in the indicator data;

[0089] For each data point in the indicator data, the normalized value is calculated using the normalization formula;

[0090] The calculated normalized value is used to replace the original data to obtain a normalized data set.

[0091] S213, setting an initial projection direction vector for each indicator data based on a heuristic method, and multiplying the initial weight value by the normalized indicator data to obtain an indicator projection value;

[0092] Specifically, setting the initial projection direction vector for each indicator data based on the heuristic method includes the following steps:

[0093] Determine the target dimension to project the data onto; if the target is 2D visualization, then the projection dimension is 2; if it is 3D visualization, then the projection dimension is 3, and so on;

[0094] For each target dimension, a unit vector is randomly generated for each target dimension using a random generation method.

[0095] S214. According to the obtained indicator projection value of each indicator data, the standard deviation value and the density value of the indicator projection value are calculated respectively.

[0096] Specifically, according to the indicator projection value of each indicator data, the calculation formulas for calculating the standard deviation value and density value of the indicator projection value are as follows:

[0097] ;

[0098] ;

[0099] Where V H Represents the standard deviation of the index projection value of the i-th index data in the product order data;

[0100] H i Represents the index projection value of the i-th index data in the product order data;

[0101] Represents the average value of the index projection values ​​of n index data in the product order data;

[0102] G H The density value representing the index projection value of the i-th index data in the product order data;

[0103] K represents the local width parameter; the value is 0.1V H ;

[0104] d i,j Represents the difference between the i-th index data in the product order data and the j-th index data in the product order data;

[0105] C(Kd i,j) represents a unit step function, if Kd i,j =0, the value is 1, otherwise the value is 0.

[0106] S22, optimizing the projection index parameters through the projection index optimization algorithm (genetic simulated annealing algorithm), and calculating the importance of each product order;

[0107] It should be noted that the main idea of ​​the projection index optimization algorithm is the genetic algorithm, which is an optimization algorithm that simulates the biological evolution process. It searches for the global optimal solution in the search space through operations such as selection, crossover and mutation.

[0108] As a preferred implementation, the projection index parameters are optimized by a projection index optimization algorithm (genetic simulated annealing algorithm), and the importance of each product order is calculated, including the following steps:

[0109] S221, randomly initialize the projection direction vector of each indicator data to generate an initial projection direction vector population;

[0110] It should be noted that in optimization algorithms or heuristic searches, it is often necessary to randomly initialize a set of candidate solutions as starting points, which also applies to randomly initializing the projection direction vector in a multidimensional space. The projection direction vector can be regarded as a point in a high-dimensional space, and each component of it represents the direction of an indicator in the projection space.

[0111] S222, for each individual in the initial projection direction vector population, calculate its fitness value, where the fitness value is the product of the standard deviation value and the density value;

[0112] S223, performing selection, crossover and mutation operations on the projection direction vector population according to the fitness value to generate a new projection direction vector population;

[0113] It should be noted that individuals in the population are selected based on their fitness values. Common selection strategies include roulette selection, tournament selection, etc. The goal of the selection operation is to retain individuals with higher fitness and eliminate individuals with lower fitness, thereby guiding the population to evolve in a better direction;

[0114] Randomly select two parent individuals for crossover operation to generate new offspring individuals. The crossover operation simulates the gene recombination process in biology. It generates new gene combinations (new projection direction vectors) by exchanging or combining some genes of the parent individuals (some components of the projection direction vector in this case). There are many specific ways to implement the crossover operation, such as single-point crossover, multi-point crossover, uniform crossover, etc. In the optimization of the projection direction vector, linear combination or rotation between vectors can be used to achieve crossover.

[0115] Perform mutation operations on individuals in the population to increase the diversity of the population. The mutation operation simulates the gene mutation process in biology. It introduces new gene combinations (new projection direction vectors) by randomly changing certain genes of individuals (in this case, certain components of the projection direction vector). There are also many specific ways to implement the mutation operation, such as randomly changing a component of a vector, randomly rotating a vector, etc. The purpose of the mutation operation is to introduce new search directions while maintaining the diversity of the population, which helps to find better solutions.

[0116] S224, repeating steps S222 to S223 until the preset number of iterations is met, selecting the projection direction vector with the highest fitness as the optimal projection direction vector, and calculating the index projection value of each product order in this direction;

[0117] S225. Use the index projection value of each product order as the importance of each product order.

[0118] S23. Construct an initial directed graph, and prioritize the product orders using a graph theory algorithm according to the importance of each product order.

[0119] As a preferred implementation, constructing an initial directed graph and prioritizing the order of goods by using a graph theory algorithm according to the importance of each order of goods includes the following steps:

[0120] S231, randomly selecting two groups of product orders as nodes, and establishing edges according to the importance between the two groups of product orders to form an initial directed graph;

[0121] It should be noted that, according to the importance between the two groups of product orders, an edge is established in the directed graph from one group of orders to the other group of orders; the direction of the edge is usually from the group with lower importance to the group with higher importance, indicating a dependency or priority.

[0122] S232, inserting new product order nodes into the initial directed graph one by one according to the importance of the product orders;

[0123] S233, analyzing whether there is a dependency relationship between the new product order node and the product order node in the initial directed graph;

[0124] It should be noted that analyzing whether a new product order node has a dependency relationship with a product order node in the initial directed graph usually involves the business logic and relevance between orders, which usually involves the following aspects:

[0125] Dependencies on raw materials or components: One order may depend on raw materials or components being provided by another order.

[0126] Dependencies in the production process: Certain orders may require certain specific production steps or stages to be completed first, which may be represented by different orders.

[0127] Time or sequence dependencies: Orders may be ordered in time, i.e. one order must be completed after another.

[0128] S234. If there is a dependency relationship, the new product order node is connected to the product order node in the initial directed graph to form an edge; if there is no dependency relationship, the new product order node is compared with the product order node in the initial directed graph in terms of importance, and the new product order node is connected to the product order node in the initial directed graph to form an edge according to the comparison result;

[0129] It should be noted that by checking whether the new product order has a dependency relationship with any existing order in the initial directed graph, if the new product order depends on an existing order, a directed edge is established from the existing order to the new order; if multiple orders depend on one product order, the orders are connected according to their importance;

[0130] If the new product order has no direct dependency with any existing order in the initial directed graph, edges need to be established based on importance.

[0131] S235. Repeat steps S232-S234 until all nodes are inserted into the directed graph to obtain a complete directed graph of goods orders, and use the topological sorting method in the graph theory algorithm to prioritize the formed directed graph of goods orders.

[0132] Specifically, using the topological sorting method in the graph theory algorithm to prioritize the formed directed graph of product orders includes the following steps:

[0133] Select a node with an in-degree of 0. The in-degree is the number of edges pointing to a node. In the case of product orders, a node with an in-degree of 0 indicates an order with no pre-dependencies and can be processed immediately. Select a node with an in-degree of 0 as the starting point for topological sorting.

[0134] Remove the selected node from the graph and delete all edges from the node; this means that the order has been processed and no longer affects the processing order of subsequent orders;

[0135] Since a node and its outgoing edges are removed, the in-degree of some nodes may become 0; update the in-degree information of all nodes;

[0136] Continue to select nodes with in-degree 0, remove them and their out-edges, and update the in-degrees of the remaining nodes until there are no nodes in the graph;

[0137] Output the sequence of all nodes in the order in which they are removed. This sequence is the priority order of the goods order.

[0138] S3. Based on the priority sorting results, the storage location of the goods is optimized through the constructed warehouse digital twin model to determine the optimal storage location for each product;

[0139] As a preferred implementation, based on the priority sorting results, the storage location of the goods is optimized through the constructed warehouse digital twin model, and determining the optimal storage location for each product includes the following steps:

[0140] S31. Integrate the priority sorting results into the warehouse digital twin model and simulate the storage locations of different goods;

[0141] S32, establishing a goods location storage model with minimizing the goods picking path as the objective function;

[0142] It should be noted that the objective function reflects the minimization of the total picking path length, which can be expressed as the minimization of the sum of the path lengths of all goods from the starting point to their storage location and then back to the starting point;

[0143] When establishing a product location storage model, constraints need to be set. Constraints may include: storage rule constraints, accessibility constraints, and security and compliance constraints;

[0144] Among them, storage rules constrain that certain goods may require a specific storage environment (such as temperature control, height restrictions, etc.).

[0145] Accessibility constraints ensure that the locations of all items can be easily accessed by picking machines or personnel.

[0146] Safety and compliance constraints are to comply with safety specifications and regulations regarding the storage of goods.

[0147] S33. Solve the goods location storage model through a location solving algorithm to obtain the optimal storage location of the goods.

[0148] As a preferred implementation, solving the goods location storage model by a location solving algorithm to obtain the optimal storage location of the goods includes the following steps:

[0149] S331. Randomly generate a storage plan for the location of goods, and simulate the picking path of each product based on the warehouse digital twin model;

[0150] S332, calculating the picking path lengths of all goods in each storage solution, and sorting the goods according to the picking path lengths to obtain a path sorting result;

[0151] As a preferred implementation, calculating the priority loss value between the path sorting result and the priority sorting result includes the following steps:

[0152] S3321. Determine the sorting position of each product according to the obtained path sorting result and priority sorting result;

[0153] S3322. For each product, calculate the difference in its sorting position in the two sorting results to obtain a sorting difference value;

[0154] It should be noted that, for each product, the absolute difference between its position in the priority sorting and the path sorting is calculated as the sorting difference value.

[0155] S3323. Add the sorting difference values ​​of all goods, and use the added result as the priority loss value between the path sorting result and the priority sorting result.

[0156] S333, calculating the priority loss value between the path sorting result and the priority sorting result;

[0157] S334. If the priority loss value is lower than the preset threshold, the goods storage location corresponding to the path sorting result is used as the optimal storage location for the goods; otherwise, the storage plan is improved using an iterative optimization algorithm to generate a new storage plan, and steps S332-S334 are repeated until the preset stop condition is met.

[0158] In order to more specifically illustrate the application of the above steps in solving the goods location storage model, the specific implementation method is as follows:

[0159] Suppose there is a warehouse with three shelves, each shelf has three cargo spaces, namely A1, A2, A3; B1, B2, B3; C1, C2, C3. There are four kinds of goods P1, P2, P3, P4, and their priorities (based on certain rules such as turnover rate) are 3, 1, 4, 2 respectively. The goal is to find a goods storage plan that makes the total path of goods picking shortest and satisfies the priorities of goods as much as possible.

[0160] Randomly generate a storage plan for the product locations, for example: P1 is placed in A1, P2 is placed in B2, P3 is placed in C3, and P4 is placed in A3;

[0161] Use the warehouse digital twin model to simulate the picking path of each product. Assume that the picking path length of all products in each storage solution has been calculated, and sort the products according to the picking path length to obtain the path sorting result. For example:

[0162] The path length of P3 (C3) is 8 units; the path length of P1 (A1) is 10 units;

[0163] The path length of P4 (A3) is 12 units; the path length of P2 (B2) is 15 units;

[0164] The path sorting result is: P3, P1, P4, P2.

[0165] Then calculate the priority loss value between the path sorting result and the priority sorting result (P3, P1, P4, P2). For example:

[0166] P3 is ranked first in the path sorting and also ranked first in the priority sorting;

[0167] P1 is ranked second in the path sorting and also ranked second in the priority sorting;

[0168] P4 is the 3rd in path sorting and the 3rd in priority sorting;

[0169] P2 is 4th in the path order and 4th in the priority order;

[0170] The sorting difference values ​​are:

[0171] The sorting difference value of P3 = |1-1| = 0; the sorting difference value of P1 = |2-2| = 0;

[0172] The sorting difference value of P4 = |3-3| = 0; the sorting difference value of P2 = |4-4| = 0;

[0173] Then the total priority loss value = 0+0+0+0=0.

[0174] In this example, the priority loss value is 0, indicating that the randomly generated solution is already a perfect solution because it satisfies both the shortest path and the priority sorting.

[0175] S4. Dynamically adjust the placement of goods in the warehouse based on the optimal storage location of each product.

[0176] like Figure 2 As shown, according to another embodiment of the present invention, a logistics warehousing management system based on digital twin is provided, and the logistics warehousing management system based on digital twin includes: a model building module 1, a priority analysis module 2, a storage location determination module 3 and a storage location adjustment module 4, and the model building module 1, the priority analysis module 2, the storage location determination module 3 and the storage location adjustment module 4 are connected in sequence;

[0177] Model building module, used to obtain warehouse logistics data and structural layout data and build a warehouse digital twin model;

[0178] Priority analysis module 2, used to obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection tracking algorithm, and prioritize the goods orders through the directed graph analysis method;

[0179] The storage location determination module 3 is used to optimize the storage location of the goods based on the priority sorting result through the constructed warehouse digital twin model to determine the optimal storage location for each product;

[0180] The storage position adjustment module 4 is used to dynamically adjust the placement of goods in the warehouse according to the optimal storage position of each product.

[0181] Figure 3 An embodiment of a computer device of the present invention is shown. The computer device may be a server, and the computer device includes a processor, a memory, and a network interface connected via a system bus. 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 static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps in the above method embodiment are implemented.

[0182] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention 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.

[0183] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiment when executing the computer program.

[0184] In addition, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0185] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention realizes accurate simulation and prediction of actual warehousing by constructing a warehousing digital twin model, which helps managers to discover potential problems in advance and optimize warehousing processes. The digital twin technology can optimize warehousing resource allocation and reduce unnecessary waste; by simulating and optimizing the warehousing process, the investment in manpower, material and financial resources can be reduced, thereby reducing operating costs; by optimizing the storage location and warehousing process of goods, the speed and accuracy of goods out of the warehouse can be improved, thereby shortening customer waiting time and improving customer satisfaction; the present invention uses the range transformation method to pre-process and normalize the data, which can eliminate the dimensional differences between different indicators, making the comparison between different indicators more meaningful. Through the projection indicator optimization algorithm, the projection direction vector can be automatically adjusted to find the optimal projection direction, so that the projection indicator parameters can more accurately reflect the importance of the goods order, and the importance of the goods order is calculated in combination with the standard deviation value and the density value, which can comprehensively consider the discrete degree and distribution of the order data. The characteristics of the distribution make the calculation of importance more comprehensive and reasonable, construct an initial directed graph and insert nodes and connect edges based on the importance of the goods order, which can intuitively display the dependency and priority relationship between the goods orders, and use the topological sorting method in the graph theory algorithm to prioritize the directed graph of the goods order, which can ensure the rationality and accuracy of the sorting result; the present invention integrates the priority sorting results into the warehouse digital twin model and simulates the storage locations of different goods. It can accurately predict and evaluate the warehousing efficiency under different storage strategies, and takes minimizing the goods picking path as the objective function, ensuring that while meeting the priority requirements, the storage location of the goods can maximize the reduction of picking time and distance, thereby significantly improving the efficiency and response speed of warehousing operations, and the optimized storage location of the goods reduces unnecessary handling and movement, reduces the warehousing operation costs, including labor costs, time costs and equipment loss costs, etc. At the same time, by reducing the picking time and distance, it can also reduce the fines and return costs caused by delayed delivery.

[0186] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0187] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A logistics warehousing management method based on digital twins, characterized in that: The digital twin-based logistics warehousing management method includes the following steps: S1. Obtain the logistics data and structural layout data of the warehouse and build a digital twin model of the warehouse; S2. Obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection tracking algorithm, and prioritize the goods orders through the directed graph analysis method; S3. Based on the priority sorting results, the storage location of the goods is optimized through the constructed warehouse digital twin model to determine the optimal storage location for each product; S4. Dynamically adjust the placement of goods in the warehouse according to the optimal storage location of each product; The method of obtaining the order data of goods in the warehouse, analyzing the importance of the goods orders by the projection pursuit algorithm, and prioritizing the goods orders by the directed graph analysis method includes the following steps: S21, preprocessing the order data of goods in the warehouse by using the range transformation method and calculating the projection index parameters, wherein the projection index parameters include the standard deviation value and the density value; S22, optimizing the projection index parameters through the projection index optimization algorithm, and calculating the importance of each product order; S23, constructing an initial directed graph, and prioritizing the product orders by using a graph theory algorithm according to the importance of each product order; The method of using the range transformation method to pre-process the order data of goods in the warehouse and calculate the projection index parameters includes the following steps: S211, performing data cleaning on each indicator data in the goods order data in the warehouse; S212, using the range transformation method to normalize each indicator data after data cleaning; S213, setting an initial projection direction vector for each indicator data based on a heuristic method, and multiplying the initial weight value by the normalized indicator data to obtain an indicator projection value; S214, according to the obtained indicator projection value of each indicator data, respectively calculating the standard deviation value and the density value of the indicator projection value; The calculation formulas for respectively calculating the standard deviation and density of the index projection values ​​obtained for each index data are as follows: ； ； Where V H Represents the standard deviation of the index projection value of the i-th index data in the product order data; H i Represents the index projection value of the i-th index data in the product order data; Represents the average value of the index projection values ​​of n index data in the product order data; G H The density value representing the index projection value of the i-th index data in the product order data; K represents the local width parameter; d i,j Represents the difference between the i-th index data in the product order data and the j-th index data in the product order data; C(Kd i,j ) represents the unit step function; The method of optimizing the projection index parameters by using the projection index optimization algorithm and calculating the importance of each product order includes the following steps: S221, randomly initialize the projection direction vector of each indicator data to generate an initial projection direction vector population; S222, for each individual in the initial projection direction vector population, calculating its fitness value, wherein the fitness value is the product of the standard deviation value and the density value; S223, performing selection, crossover and mutation operations on the projection direction vector population according to the fitness value to generate a new projection direction vector population; S224, repeating steps S222 to S223 until the preset number of iterations is met, selecting the projection direction vector with the highest fitness as the optimal projection direction vector, and calculating the index projection value of each product order in this direction; S225, taking the index projection value of each product order as the importance of each product order; The construction of the initial directed graph and prioritizing the order of goods by using a graph theory algorithm according to the importance of each order of goods include the following steps: S231, randomly selecting two groups of product orders as nodes, and establishing edges according to the importance between the two groups of product orders to form an initial directed graph; S232, inserting new product order nodes into the initial directed graph one by one according to the importance of the product orders; S233, analyzing whether there is a dependency relationship between the new product order node and the product order node in the initial directed graph; S234. If there is a dependency relationship, the new product order node is connected to the product order node in the initial directed graph to form an edge; if there is no dependency relationship, the new product order node is compared with the product order node in the initial directed graph in terms of importance, and the new product order node is connected to the product order node in the initial directed graph to form an edge according to the comparison result; S235, repeating steps S232-S234 until all nodes are inserted into the directed graph, obtaining a complete product order directed graph, and using a topological sorting method in a graph theory algorithm to prioritize the formed product order directed graph; Based on the priority sorting results, the storage location of the goods is optimized through the constructed warehouse digital twin model, and determining the optimal storage location of each product includes the following steps: S31. Integrate the priority sorting results into the warehouse digital twin model and simulate the storage locations of different goods; S32, establishing a goods location storage model with minimizing the goods picking path as the objective function; S33, solving the goods location storage model by using a location solving algorithm to obtain the optimal storage location of the goods; Solving the goods location storage model by using a location solving algorithm to obtain the optimal storage location of the goods includes the following steps: S331. Randomly generate a storage plan for the location of goods, and simulate the picking path of each product based on the warehouse digital twin model; S332, calculating the picking path lengths of all goods in each storage solution, and sorting the goods according to the picking path lengths to obtain a path sorting result; S333, calculating the priority loss value between the path sorting result and the priority sorting result; S334. If the priority loss value is lower than a preset threshold, the goods storage location corresponding to the path sorting result is used as the optimal storage location for the goods; otherwise, the storage plan is improved using an iterative optimization algorithm to generate a new storage plan, and steps S332-S334 are repeated until a preset stop condition is met; The calculation of the priority loss value between the path sorting result and the priority sorting result comprises the following steps: S3321. Determine the sorting position of each product according to the obtained path sorting result and priority sorting result; S3322. For each product, calculate the difference in its sorting position in the two sorting results to obtain a sorting difference value; S3323. Add the sorting difference values ​​of all goods, and use the added result as the priority loss value between the path sorting result and the priority sorting result.

2. A logistics warehousing management system based on digital twins, used to implement the logistics warehousing management method based on digital twins described in claim 1, characterized in that: The digital twin-based logistics warehousing management system includes: a model building module, a priority analysis module, a storage location determination module and a storage location adjustment module, and the model building module, the priority analysis module, the storage location determination module and the storage location adjustment module are connected in sequence; The model building module is used to obtain the logistics data and structural layout data of the warehouse and build a warehouse digital twin model; The priority analysis module is used to obtain the order data of goods in the warehouse, analyze the importance of the goods orders through the projection pursuit algorithm, and prioritize the goods orders through the directed graph analysis method; The storage location determination module is used to optimize the storage location of the goods based on the priority sorting result through the constructed warehouse digital twin model to determine the optimal storage location for each product; The storage position adjustment module is used to dynamically adjust the placement of goods in the warehouse according to the optimal storage position of each product.

Images