Logistics digital warehouse management system and method

Through enhanced feedforward neural network and adaptive position optimization algorithm to calculate the optimal storage location, combined with deep-first search and reinforcement learning algorithm optimization path, the problems of insufficient data processing capabilities and imbalanced path optimization in traditional warehousing management are solved, and the problem of insufficient data processing capabilities and imbalanced path optimization in traditional warehousing management is improved. The scientific nature and operational efficiency of the warehouse.

CN118644187BActive Publication Date: 2024-11-01ZHANGJIAGANG FREE TRADE ZONE CHANGJIANG INT PORT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411124115.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-01
Estimated Expiration
2044-08-16

AI Technical Summary

Technical Problem

In traditional warehousing management systems, the data processing capability is insufficient, and the selection of items storage locations lacks scientificity and rationality, resulting in insufficient space utilization and low access efficiency, and lack of balance mechanism for path optimization algorithms, making it difficult to effectively analyze warehousing data and optimize item storage paths. .

Method used

The optimal storage location is calculated using enhanced feedforward neural network and adaptive position optimization algorithm, combined with the reinforcement learning algorithm of depth-first search and optimization, select the optimal path, optimize the storage location and path through nonlinear adjustment and correction coefficients, and comprehensively consider items. Weight, frequency, compatibility, stacking limitations, value and environmental factors.

Benefits of technology

It significantly improves the warehouse space utilization and management efficiency, optimizes the placement and storage paths of items, reduces time and energy consumption, and improves the warehouse operation efficiency and overall intelligence level.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118644187B_ABST
    Figure CN118644187B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of digital logistics warehouse management systems, and in particular to digital logistics warehouse management systems and methods. The contents include: acquiring and preprocessing data inside and outside the warehouse, using an enhanced feedforward neural network to analyze the preprocessed data, and outputting the best storage location parameters; based on the best storage location parameters, calculating the best storage location through an adaptive location optimization algorithm; based on the best storage location, calculating candidate paths; and selecting candidate paths by using an optimized reinforcement learning algorithm to obtain the optimal path. The present invention solves the technical problems that traditional data processing and analysis methods have insufficient processing capabilities and are difficult to effectively analyze warehouse data; in existing warehouse management systems, the selection of item storage locations lacks scientificity and rationality, and the item access paths are not optimized enough; and the existing path optimization algorithms lack an effective balancing mechanism when considering path length, time consumption, energy consumption, and congestion levels.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of digital logistics warehouse management systems, and in particular to digital logistics warehouse management systems and methods. Background Art

[0002] With the rapid development of globalization and e-commerce, the importance of logistics and supply chain management has become increasingly prominent. Warehousing, as a key link in the logistics chain, is a place for the storage and management of goods, and is also an important guarantee for the efficiency of supply chain operations. However, the traditional warehouse management method relies on manual operation and paper records, and has problems such as low operating efficiency, poor data accuracy, information lag, and chaotic inventory management. There is an urgent need for an efficient, accurate, and real-time digital warehouse management system and method to improve the overall level of warehouse management. The logistics digital warehouse management system and method demonstrate unique technical advantages, realize the automation, intelligence, and informatization of warehouse management, effectively solve many problems in traditional warehouse management, and improve the efficiency and accuracy of warehouse management. With the continuous development of technology and the deepening of its application, the digital warehouse management system will become an important tool for modern logistics and supply chain management, providing strong support for the development of enterprises.

[0003] my country's patent application number: CN202310154667.6, publication date: 2023.03.28, discloses a logistics warehousing management method and system based on big data, which generates logistics simulation prediction data by acquiring logistics area big data and performing cargo data analysis and cargo order prediction, and performs logistics resource layout analysis based on logistics simulation prediction data, obtains multiple regional resource allocation data, and performs multi-region data fusion and difference analysis based on regional resource allocation data, and obtains regional logistics characteristic difference information, so as to more intuitively grasp the characteristic information of the logistics area and improve the precise regulation of the logistics process. In addition, in the event of sudden abnormalities in logistics, resource allocation and regulation can be carried out according to the characteristic differences of the logistics area, thereby improving the robustness of the logistics system and reducing costs and increasing efficiency for logistics companies.

[0004] However, the above technologies have at least the following technical problems: traditional data processing and analysis methods have insufficient processing capabilities and it is difficult to effectively analyze warehouse data; in existing warehouse management systems, the storage location of items relies on manual experience, and the selection of item storage locations lacks scientificity and rationality, which will lead to insufficient space utilization and unreasonable storage; in traditional warehouse management systems, the storage and access paths of items are not optimized enough, resulting in low storage and access efficiency; existing path optimization algorithms lack an effective balancing mechanism when considering path length, time consumption, energy consumption and congestion level. Summary of the invention

[0005] The present invention provides a logistics digital warehouse management system and method to solve the problems that traditional data processing and analysis methods have insufficient processing capabilities and are difficult to effectively analyze warehouse data; in existing warehouse management systems, the storage location of items relies on manual experience, and the selection of item storage locations lacks scientificity and rationality, which will lead to insufficient space utilization and unreasonable storage; in traditional warehouse management systems, the item storage and access paths are not optimized enough, resulting in low storage and access efficiency; the existing path optimization algorithm lacks an effective balancing mechanism when considering path length, time consumption, energy consumption and congestion level.

[0006] The digital logistics warehouse management system and method of the present invention specifically include the following technical solutions:

[0007] The logistics digital warehousing management method includes the following steps:

[0008] S1. Acquire and preprocess the data inside and outside the warehouse, use an enhanced feedforward neural network to analyze the preprocessed data, and output the optimal storage location parameters; the enhanced feedforward neural network includes an input layer, three hidden layers and an output layer; the specific implementation method is as follows:

[0009] The input layer of the enhanced feedforward neural network receives the preprocessed data , the preprocessed data Input to the hidden layer for calculation and analysis; the specific formula of the first hidden layer is:

[0010] ,

[0011] in, represents the output of the first hidden layer; express Activation function; represents the weight matrix of the first hidden layer; represents the bias vector of the first hidden layer; Represents the variance of the input layer data; Used to prevent division by zero errors;

[0012] The specific formula for the second hidden layer is:

[0013] ,

[0014] in, represents the output of the second hidden layer; represents the logarithmic function; represents the cosine function; represents the weight matrix of the second hidden layer; represents the bias vector of the second hidden layer;

[0015] The specific formula of the third hidden layer is:

[0016] ,

[0017] in, represents the output of the third hidden layer; Represents the weight matrix of the third hidden layer; Represents the absolute value of the output of the second hidden layer; represents the square of the output of the second hidden layer; represents the bias vector of the third hidden layer;

[0018] The output layer receives the output from the third hidden layer and outputs the best position to store the parameters. The specific formula of the output layer is:

[0019] ,

[0020] in, Represents the output of the output layer; represents the base of natural logarithms; represents the weight matrix of the output layer; represents the sine function; Represents the bias vector of the output layer;

[0021] S2. Based on the optimal storage location parameters, the optimal storage location is calculated by an adaptive location optimization algorithm; the specific formula is:

[0022] ,

[0023] in, Indicates the best storage location; Indicates finding the minimum target position ; Indicates target locations, , Indicates the number of target positions, ; Indicates the type of item. , Indicates the number of item types; Indicates The weight of the item; Indicates The frequency of access to the items; Indicates Plant items to the target location distance; represents the compatibility adjustment factor; Indicates Items and target locations compatibility; Indicates the stack limit adjustment factor; Indicates Items at the target location Stacking limit; Indicates The value of the item; represents the value adjustment factor; Indicates environmental factors; represents the environmental factor adjustment coefficient;

[0024] S3. Based on the optimal storage location, calculate the candidate paths; select the candidate paths by using the optimized reinforcement learning algorithm to obtain the optimal path; the implementation process of the optimized reinforcement learning algorithm is as follows:

[0025] The reward value for design improvement is calculated as follows:

[0026] ,

[0027] in, represents the reward value of the improvement; Indicates a negative sign; represents the path length; Indicates the path time; represents the path energy consumption; Indicates the degree of path congestion;

[0028] An optimized reinforcement learning algorithm is designed based on the improved reward value; and candidate paths are selected by using the optimized reinforcement learning algorithm.

[0029] Preferably, the S2 specifically includes:

[0030] The adaptive location optimization algorithm optimizes the storage location based on the optimal storage location parameters through nonlinear adjustment and correction coefficients.

[0031] Preferably, the S3 specifically includes:

[0032] Based on the optimal storage location, the candidate path is calculated using the depth-first search algorithm.

[0033] Preferably, the S3 specifically includes:

[0034] The improved reward value combines factors such as path length, path time, path energy consumption and path congestion level. By changing different factors, the path with the best performance in terms of length, time, energy consumption and congestion level is selected.

[0035] The logistics digital warehouse management system includes the following parts:

[0036] Data acquisition module, data preprocessing module, enhanced feedforward neural network analysis module, adaptive position optimization algorithm module, depth-first search module, optimized reinforcement learning algorithm module;

[0037] Data acquisition module: obtains data inside and outside the warehouse, and inputs the data inside and outside the warehouse into the data preprocessing module;

[0038] Data preprocessing module: preprocesses data inside and outside the warehouse, and inputs the preprocessed data into the enhanced feedforward neural network analysis module;

[0039] Enhanced feedforward neural network analysis module: uses an enhanced feedforward neural network to analyze the preprocessed data and output the optimal storage location parameters; inputs the optimal storage location parameters into the adaptive location optimization algorithm module;

[0040] Adaptive location optimization algorithm module: receiving optimal storage location parameters, obtaining the optimal storage location through an adaptive location optimization algorithm; inputting the optimal storage location into a depth-first search module;

[0041] Depth-first search module: Based on the optimal storage location, the depth-first search algorithm is used to calculate the candidate path, and the candidate path is input into the optimized reinforcement learning algorithm module;

[0042] Optimized reinforcement learning algorithm module: The optimal path is obtained by selecting candidate paths using an optimized reinforcement learning algorithm.

[0043] The beneficial effects of the technical solution of the present invention are:

[0044] 1. By introducing an enhanced feedforward neural network based on complex nonlinear calculations, regularization technology and deep structures, it is able to conduct in-depth analysis of multi-factor data and output the optimal storage location parameters; it significantly improves the ability to handle complex relationships and provides accurate storage location optimization suggestions, thereby improving warehouse space utilization and management efficiency;

[0045] 2. The adaptive location optimization algorithm comprehensively considers the weight of items, the frequency of item storage and retrieval, the distance from the item to the target location, the compatibility of the item with the target location, the stacking limit, the value of the item and the environmental factors, and scientifically and rationally calculates the optimal storage location through nonlinear adjustment and correction coefficients; it optimizes the placement of items in the warehouse, improves the scientificity and rationality of warehouse management, reduces the time and labor cost of item storage and retrieval, and improves warehouse operation efficiency;

[0046] 3. Calculate candidate paths through the depth-first search algorithm and select them using the optimized reinforcement learning algorithm The path with the highest value is taken as the optimal path; the improved reward value comprehensively considers the path length, time consumption, energy consumption and congestion level to ensure that the selected path performs best in all aspects; it optimizes the item storage and access path, reduces the item movement time and energy consumption, improves logistics efficiency, reduces warehouse operating costs, and improves the intelligence level of the overall warehouse management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a structural diagram of the digital logistics warehousing management system described in the present invention;

[0048] Figure 2 This is a flow chart of the logistics digital warehousing management method described in the present invention. DETAILED DESCRIPTION

[0049] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0051] The specific scheme of the digital logistics warehousing management system and method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0052] See attached Figure 1 , which shows a structural diagram of a logistics digital warehousing management system provided by an embodiment of the present invention, the system includes the following parts:

[0053] Data acquisition module, data preprocessing module, enhanced feedforward neural network analysis module, adaptive position optimization algorithm module, depth-first search module, optimized reinforcement learning algorithm module;

[0054] Data acquisition module: uses RFID tags, readers, and sensor networks to acquire data inside and outside the warehouse, and inputs the acquired data inside and outside the warehouse into the data preprocessing module;

[0055] Data preprocessing module: preprocesses the data from inside and outside the warehouse of the data acquisition module, including data cleaning, denoising and standardization, obtains preprocessed data, and inputs the preprocessed data into the enhanced feedforward neural network analysis module;

[0056] Enhanced feedforward neural network analysis module: receives the preprocessed data from the data preprocessing module, uses the enhanced feedforward neural network to analyze the preprocessed data, outputs the optimal storage location parameters, and inputs the optimal storage location parameters into the adaptive location optimization algorithm module;

[0057] Adaptive location optimization algorithm module: receives the optimal storage location parameters from the enhanced feedforward neural network analysis module, calculates the optimal storage location through the adaptive location optimization algorithm; inputs the optimal storage location into the depth-first search module;

[0058] Depth-first search module: receives the best storage position from the adaptive position optimization algorithm module, calculates the candidate path based on the best storage position using the depth-first search algorithm, and inputs the candidate path into the optimized reinforcement learning algorithm module;

[0059] Optimized reinforcement learning algorithm module: receives candidate paths from the depth-first search module, selects the candidate paths by using the optimized reinforcement learning algorithm, and obtains the optimal path.

[0060] See attached Figure 2 , which shows a flow chart of a logistics digital warehousing management method provided by an embodiment of the present invention, the method comprising the following steps:

[0061] S1, obtain and preprocess the data inside and outside the warehouse, use the enhanced feedforward neural network to analyze the preprocessed data, and output the optimal storage location parameters;

[0062] Use RFID tags, readers, and sensor networks to obtain data inside and outside the warehouse , including cargo information, inbound and outbound information, and environmental data; RFID tags, readers, and sensor networks can be set up according to specific implementation scenarios and are not limited here; data inside and outside the warehouse Perform preprocessing, including data cleaning, denoising and standardization, to obtain preprocessed data ; The specific pretreatment method is a technical means well known to those skilled in the art and will not be described in detail here;

[0063] In the embodiment of the present invention, an enhanced feedforward neural network is used to process the preprocessed data. Analyze and output the optimal storage location parameters; the enhanced feedforward neural network can significantly improve the ability to process multi-factor data by introducing complex nonlinear calculations, regularization technology and deep structures;

[0064] The input layer of the enhanced feedforward neural network receives the preprocessed data , the preprocessed data Input to the hidden layer for computational analysis;

[0065] The specific formula for the first hidden layer is:

[0066] ,

[0067] in, represents the output of the first hidden layer; express Activation function, used to introduce nonlinear characteristics to ensure that the enhanced feedforward neural network can handle complex relationships; Represents the weight matrix of the first hidden layer, which is used to preprocess the data Perform linear transformation so that each input preprocessed data is connected to the hidden layer neurons; Represents the bias vector of the first hidden layer, which is used to adjust the output of the linear transformation; Represents the variance of the input layer data, which is used for normalization to ensure that the preprocessed data of different scales are balanced; Represents a constant used to prevent division by zero errors;

[0068] The specific formula for the second hidden layer is:

[0069] ,

[0070] in, represents the output of the second hidden layer; express Activation function, used to introduce nonlinear characteristics; Represents a logarithmic function, used to process the scale of the output value; Represents the cosine function, which is used to introduce periodic nonlinear transformation and increase the nonlinear expression ability of the enhanced feedforward neural network; Represents the weight matrix of the second hidden layer, connecting the output of the first hidden layer to the neurons of the second hidden layer; Represents the bias vector of the second hidden layer, which is used to adjust the linear transformation result of the second hidden layer; Represents the output of the first hidden layer, which is used for residual connection to prevent gradient disappearance and stabilize the training of enhanced feedforward neural network;

[0071] The specific formula of the third hidden layer is:

[0072] ,

[0073] in, represents the output of the third hidden layer; express Activation function, used to introduce nonlinear characteristics; Represents the weight matrix of the third hidden layer; Represents the absolute value of the output of the second hidden layer, which is used to ensure that the denominator is a positive number; Represents the square of the output of the second hidden layer, which is used to increase nonlinear characteristics; Represents the bias vector of the third hidden layer, which is used to adjust the linear transformation result of the third hidden layer; residual connection Used to avoid the vanishing gradient problem;

[0074] The output layer receives the output from the third hidden layer and outputs the best location to store the parameters;

[0075] The specific formula of the output layer is:

[0076] ,

[0077] in, Represents the output of the output layer, i.e., the optimal storage location parameter; express Activation function, used to introduce nonlinear characteristics; Represents the base of the natural logarithm, which is used in exponential functions to amplify the range of data changes; Represents the weight matrix of the output layer, connecting the output of the third hidden layer to the output layer; Represents the sine function, which is used to introduce nonlinear transformations and process periodic data; represents the output of the third hidden layer; Represents the bias vector of the output layer, which is used to adjust the final linear transformation result;

[0078] The input layer of the enhanced feedforward neural network receives the preprocessed data ,The output layer outputs the parameters of the best storage location, such as item weight, item access frequency, item to target location, item compatibility with target location, stacking restrictions, item value, and environmental factors, which can be used for the calculation of the best storage location.

[0079] S2. Based on the optimal storage location parameters, the optimal storage location is calculated by an adaptive location optimization algorithm;

[0080] In the embodiment of the present invention, based on the optimal storage location parameters, the optimal storage location is calculated by an adaptive location optimization algorithm; the adaptive location optimization algorithm is a method based on comprehensive consideration of multiple factors, used to calculate the optimal storage location of items in the warehouse, the adaptive location optimization algorithm combines the weight of the item, the frequency of access to the item, the distance from the item to the target location, the compatibility of the item with the target location, the stacking limit, the value of the item, and environmental factors, and through nonlinear adjustment and correction coefficients, the storage location selection is scientific and reasonable;

[0081] The specific formula for calculating the optimal storage location through the adaptive location optimization algorithm is:

[0082] ,

[0083] in, Indicates the best storage location; Indicates finding the minimum target position ; Indicates target locations, , Indicates the number of target positions, Indicates target locations, ; Indicates the type of item. , Indicates the number of item types; Indicates The weight of the items reflects the impact of the weight of the items on the choice of storage location; Indicates The frequency of access to the items; Indicates Plant items to the target location distance; represents a nonlinear adjustment term; Indicates the compatibility adjustment factor, adjusting the item and the target position The weight of compatibility; Indicates the compatibility of the item with the target location, reflecting the The degree of compatibility between the items and the candidate storage locations; Indicates the stacking limit adjustment coefficient, which adjusts the weight of the item stacking limit; Indicates Items at the target location Stacking limit; Indicates The value of the item; Represents the value adjustment coefficient, which adjusts the impact of the item value on the target position; Represents environmental factors, considering the environmental conditions of each target location in the warehouse, The impact of the choice of storage location for various items; It represents the environmental factor adjustment coefficient, which regulates the impact of environmental factors on the target position;

[0084] The optimal storage location is calculated through an adaptive location optimization algorithm to improve warehouse management efficiency and the convenience of storing and accessing items.

[0085] S3. Based on the optimal storage location, calculate the candidate paths; select the candidate paths by using the optimized reinforcement learning algorithm to obtain the optimal path.

[0086] In the embodiment of the present invention, based on the optimal storage location, a depth-first search algorithm is used to calculate the candidate path , Indicates the number of candidate paths. Indicates candidate paths, ; The depth-first search algorithm is a technical means well known to those skilled in the art and will not be described in detail here;

[0087] An optimized reinforcement learning algorithm is used to select candidate paths; the optimized reinforcement learning algorithm is optimized by designing improved reward values;

[0088] The improved reward value comprehensively considers factors such as path length, path time, path energy consumption, and path congestion. By appropriately transforming different factors, the influence of each factor is smoothed, ensuring that the selected path performs best in terms of length, time, energy consumption, and congestion.

[0089] The specific calculation formula of the improved reward value is:

[0090] ,

[0091] in, Represents the improved reward value, which is used to measure the quality of the candidate path; Represents a negative sign, ensuring that the reward value is negative, so that the optimization goal is to minimize the negative penalty, that is, to maximize the positive reward; represents the path length; Indicates the logarithmic transformation of the path length. The logarithmic function smoothes the effect of the path length. Make sure the input value of the logarithm function is always greater than zero; Indicates the path time; Indicates that the path time is transformed by square root, which mitigates the growth rate of time. represents the path energy consumption; Indicates that the path energy consumption is transformed into a power function, which can smooth the impact of energy consumption; Indicates the degree of path congestion; It represents the square transformation of the path congestion degree. The square function increases the penalty of congestion degree on reward, so as to select unobstructed paths.

[0092] Improvement-based reward value Get optimized reinforcement learning algorithm;

[0093] The specific formula of the optimized reinforcement learning algorithm is:

[0094] ,

[0095] in, Indicates in status Take action The value function of Represents the learning rate, ranging from 0 to 1, which controls the proportion of new values ​​in the update; represents the discount factor; represents the reward value of the improvement; Indicates the next state All possible actions The largest Value, indicating that in the next state Next, choose the best action The best value that can be obtained; Indicates the next state, indicating that the action is being executed The new state to which the system is transferred; Indicates the action in the next state, indicating that in the state All possible actions that can be taken;

[0096] By using an optimized reinforcement learning algorithm to select candidate paths, The path with the highest value is taken as the optimal path. The specific formula is:

[0097] ,

[0098] in, represents the optimal path; Used to find all candidate paths such that The path with the largest value; Indicates Candidate Paths of value; Indicates candidate paths.

[0099] In summary, the logistics digital warehousing management system and method are completed.

[0100] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0101] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0102] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A digital logistics warehousing management method, characterized in that: The following steps are involved: S1. Acquire and preprocess the data inside and outside the warehouse, use an enhanced feedforward neural network to analyze the preprocessed data, and output the optimal storage location parameters; the enhanced feedforward neural network includes an input layer, three hidden layers and an output layer; the specific implementation method is as follows: The input layer of the enhanced feedforward neural network receives the preprocessed data , the preprocessed data Input to the hidden layer for calculation and analysis; the specific formula of the first hidden layer is: , in, represents the output of the first hidden layer; express Activation function; represents the weight matrix of the first hidden layer; represents the bias vector of the first hidden layer; Represents the variance of the input layer data; Used to prevent division by zero errors; The specific formula for the second hidden layer is: , in, represents the output of the second hidden layer; represents the logarithmic function; represents the cosine function; represents the weight matrix of the second hidden layer; represents the bias vector of the second hidden layer; The specific formula of the third hidden layer is: , in, represents the output of the third hidden layer; Represents the weight matrix of the third hidden layer; Represents the absolute value of the output of the second hidden layer; represents the square of the output of the second hidden layer; represents the bias vector of the third hidden layer; The output layer receives the output from the third hidden layer and outputs the best position to store the parameters. The specific formula of the output layer is: , in, Represents the output of the output layer; represents the base of natural logarithms; represents the weight matrix of the output layer; represents the sine function; Represents the bias vector of the output layer; S2. Based on the optimal storage location parameters, the optimal storage location is calculated by an adaptive location optimization algorithm; the specific formula is: , in, Indicates the best storage location; Indicates finding the minimum target position ; Indicates target locations, , Indicates the number of target positions, ; Indicates the type of item. , Indicates the number of item types; Indicates The weight of the item; Indicates The frequency of access to the items; Indicates Plant items to the target location distance; represents the compatibility adjustment factor; Indicates Items and target locations compatibility; Indicates the stack limit adjustment factor; Indicates Items at the target location Stacking limit; Indicates The value of the item; represents the value adjustment factor; Indicates environmental factors; represents the environmental factor adjustment coefficient; S3. Based on the optimal storage location, calculate the candidate paths; select the candidate paths by using the optimized reinforcement learning algorithm to obtain the optimal path; the implementation process of the optimized reinforcement learning algorithm is as follows: The reward value for design improvement is calculated as follows: , in, represents the reward value of the improvement; Indicates a negative sign; represents the path length; Indicates the path time; represents the path energy consumption; Indicates the degree of path congestion; An optimized reinforcement learning algorithm is designed based on the improved reward value; candidate paths are selected by using the optimized reinforcement learning algorithm; the specific formula of the optimized reinforcement learning algorithm is: , in, Indicates in status Take action The value function of represents the learning rate; represents the discount factor; Indicates the next state All possible actions The largest value.

2. The digital warehousing management method for logistics according to claim 1 is characterized in that: The S2 specifically includes: The adaptive location optimization algorithm optimizes the storage location based on the optimal storage location parameters through nonlinear adjustment and correction coefficients.

3. The logistics digital warehousing management method according to claim 1 is characterized in that: The S3 specifically includes: Based on the optimal storage location, the candidate path is calculated using the depth-first search algorithm.

4. The logistics digital warehousing management method according to claim 1 is characterized in that: The S3 specifically includes: The improved reward value combines factors such as path length, path time, path energy consumption and path congestion level. By changing different factors, the path with the best performance in terms of length, time, energy consumption and congestion level is selected.

5. A digital logistics warehouse management system, applied to the digital logistics warehouse management method as claimed in claim 1, characterized in that: Includes the following sections: Data acquisition module, data preprocessing module, enhanced feedforward neural network analysis module, adaptive position optimization algorithm module, depth-first search module, optimized reinforcement learning algorithm module; Data acquisition module: obtains data inside and outside the warehouse, and inputs the data inside and outside the warehouse into the data preprocessing module; Data preprocessing module: preprocesses data inside and outside the warehouse, and inputs the preprocessed data into the enhanced feedforward neural network analysis module; Enhanced feedforward neural network analysis module: Uses enhanced feedforward neural network to analyze preprocessed data and outputs optimal storage location parameters; Inputting the optimal storage location parameters into the adaptive location optimization algorithm module; Adaptive location optimization algorithm module: receiving optimal storage location parameters, obtaining the optimal storage location through an adaptive location optimization algorithm; inputting the optimal storage location into a depth-first search module; Depth-first search module: Based on the optimal storage location, the depth-first search algorithm is used to calculate the candidate path, and the candidate path is input into the optimized reinforcement learning algorithm module; Optimized reinforcement learning algorithm module: The optimal path is obtained by selecting candidate paths using an optimized reinforcement learning algorithm.