Industrial stereoscopic warehouse AI intelligent management method and system

Abstract

The invention relates to the technical field of industrial stereoscopic warehouse management, in particular to an industrial stereoscopic warehouse AI intelligent management method and system, and the system comprises a data collection module, a parameter configuration module, a path optimization module, a prediction analysis module and a task distribution module. Through path optimization, cargo storage information prediction is carried out through an LSTM time sequence prediction model, and equipment task allocation is realized, so that whole-process intelligent management of warehouse cargo storage, sorting and transportation is completed. According to the method, the resource utilization efficiency can be optimized, the real-time response capability and the multi-task concurrent processing capability of the system are improved, and the efficient and intelligent warehouse management requirements of the modern industry are met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent warehousing and artificial intelligence technology, and specifically to an AI intelligent management method and system for an industrial high-bay warehouse. Background Art

[0002] Warehouse management, also known as storage management, refers to the effective control of the receipt, delivery, and inventory of stored goods. Its purpose is to ensure the integrity of stored goods and the normal operation of production and business activities for enterprises. On this basis, the activity status of various types of goods is recorded and categorized, and the status of stored goods in terms of quantity and quality, as well as the geographical location, department, order affiliation, and degree of storage dispersion are expressed in clear charts. This is a comprehensive management form. Warehouse management is related to logistics and transportation. Generally, after sorting, various types of goods stored in the warehouse are subject to a variety of different logistics and transportation methods, including local express delivery, long-distance express delivery, dedicated vehicle and person delivery, global air transport, and global sea transport. Therefore, it is necessary to introduce a dynamic management mechanism based on different logistics and transportation methods into warehouse management to ensure that limited logistics and transportation resources can match the changing dynamics of goods.

[0003] Patent publication number CN115034391B is a one-stop model compression system for intelligent detection of welding defects. This patent realizes lightweight compression of quality defect detection models through modules such as model warehouse management, compression algorithm configuration, and resource scheduling, thereby improving the reasoning efficiency of the model on the edge side. Although this technical solution is innovative in model compression and resource scheduling, its application scenarios are mainly concentrated in the field of welding defect detection, which is different from the management needs of industrial warehouses. Specifically, this solution does not involve the intelligent management of the entire process of goods storage, sorting, transportation, etc. in industrial warehouses, nor does it propose solutions for multi-dimensional data fusion and complex task scheduling in warehouses. Therefore, it is difficult to directly apply it to the AI ​​intelligent management scenario of industrial warehouses.

[0004] However, the above-mentioned existing technologies still have room for improvement in terms of overall intelligent management of industrial warehouses, adaptability to complex scenarios, and full-process optimization. For example, in dynamic inventory optimization, there is a lack of comprehensive consideration of cargo distribution density, storage priority, and turnover rate; in equipment collaborative scheduling, path planning algorithms and task allocation strategies are not fully integrated to achieve efficient resource utilization; and in multidimensional data analysis, the application of time series prediction models is relatively limited. These factors make it difficult for existing technologies to fully meet the needs of modern industry for efficient and intelligent warehouse management. To this end, an AI intelligent management method and system for industrial warehouses is proposed to address the above-mentioned issues. Summary of the Invention

[0005] In order to solve technical problems in related fields, the present invention provides an AI intelligent management method and system for industrial high-bay warehouses. Based on comprehensive multi-dimensional data processing and dynamic task scheduling, it combines path planning algorithms, priority allocation strategies and time series prediction models to perform full-process intelligent management of cargo storage, sorting and transportation in the warehouse, thereby optimizing resource utilization efficiency and improving the system's real-time response capability.

[0006] According to one aspect of the present invention, an AI intelligent management method for an industrial stereoscopic warehouse is provided, the method comprising the following steps: collecting cargo storage information corresponding to each historical time interval of a target warehouse before a current time interval, the current time interval taking the current moment as the starting time, and the single cargo storage information corresponding to each historical time interval being the quantity, storage location distribution, turnover rate and inventory priority of each type of cargo received by the target warehouse within the historical time interval; obtaining various operating parameters of the target warehouse, the various operating parameters of the target warehouse including the maximum storage capacity of the warehouse, the number of shelf layers, the number of automated equipment and the entrance and exit throughput; dynamically optimizing the cargo storage location in the target warehouse, generating an optimal storage path, and using the optimal storage path as the path plan The output result of the planning module is collected; the LSTM time series prediction model is used to predict the quantity of each type of goods received by the target warehouse in the current time interval and their storage priority based on the storage information of each piece of goods corresponding to each historical time interval before the current time interval, the various operating parameters of the target warehouse and the duration of the time interval; based on the predicted quantity of each type of goods and their storage priority in the current time interval, the automated equipment in the target warehouse is assigned tasks and an equipment scheduling plan is generated; wherein, the storage information of each piece of goods corresponding to each historical time interval before the current time interval of the target warehouse is collected, and it also includes: the historical time intervals before the current time interval and the current time interval together constitute a complete time segment, and the duration of each time interval is equal.

[0007] According to another aspect of the present invention, an AI intelligent management system for an industrial stereoscopic warehouse is provided, the system comprising the following components: a data acquisition module for collecting storage information of goods corresponding to each historical time interval of a target warehouse before a current time interval; a parameter configuration module for obtaining various operating parameters of the target warehouse; a path optimization module connected to the data acquisition module for dynamically optimizing the storage location of goods in the target warehouse and generating an optimal storage path; a prediction analysis module connected to the data acquisition module, the parameter configuration module and the path optimization module, respectively, for using an LSTM time series prediction model to predict the goods based on the historical time intervals of the target warehouse before the current time interval. The method includes: collecting the corresponding cargo storage information of each copy, the operating parameters of the target warehouse, and the duration of the time interval to predict the quantity of each type of cargo received by the target warehouse in the current time interval and the storage priority thereof; a task allocation module is connected to the prediction analysis module, and is used to perform task allocation on the automated equipment in the target warehouse based on the predicted quantity of each type of cargo and the storage priority thereof in the current time interval, and generate an equipment scheduling plan; wherein, collecting the cargo storage information of each copy corresponding to each historical time interval before the current time interval also includes: the historical time intervals before the current time interval and the current time interval together constitute a complete time segment, and the duration of each time interval is equal.

[0008] It can be seen that the present invention has at least the following four outstanding substantial features: First: for the intelligent prediction of the quantity of various types of goods and their storage priority in the current time interval just arrived at the target warehouse, multiple basic data are targeted and screened. The multiple basic data include the various cargo storage information corresponding to each historical time interval before the current time interval and the various operating parameters of the target warehouse, thereby providing comprehensive and effective basic data for the subsequent intelligent prediction of cargo storage information in the current time interval; Second: a path optimization module is introduced to perform dynamic optimization of cargo storage locations in the target warehouse. The path optimization module generates an optimal storage path based on the spatial layout of the warehouse and the cargo storage priority, thereby completing the customized design of the storage path; Third: based on the prediction The quantity of various types of goods and their storage priority in the current time interval are used to perform task allocation on the automated equipment in the target warehouse and generate an equipment scheduling plan, so as to judge the task requirements of the automated equipment for the target warehouse in advance and realize the efficient use of limited resources; Fourthly: in each training of the LSTM time series prediction model, the quantity of various types of goods received by the target warehouse in a certain past time interval and their storage priority are used as the output content of the prediction model, and the storage information of each piece of goods corresponding to each historical time interval before the certain past time interval, the various operating parameters of the target warehouse and the duration of the time interval are used as the various input contents of the prediction model to complete this training action, thereby ensuring the effectiveness of each training.

[0009] In addition, the present invention further refines the technical solution through the following specific implementation methods: in the data acquisition module, a distributed sensor network is used to collect real-time storage information of goods in the target warehouse. The distributed sensor network is composed of multiple pressure sensors, temperature sensors and RFID tag readers installed on the shelves. The pressure sensor is used to detect the weight change of the goods, the temperature sensor is used to monitor the storage environment conditions of the goods, and the RFID tag reader is used to record the unique identification information of the goods; in the path optimization module, a two-dimensional grid map is constructed based on the spatial layout diagram of the target warehouse. Each grid unit in the two-dimensional grid map corresponds to a storage area in the warehouse. By calculating the minimum cost path from the starting point to the end point, the minimum cost path is obtained. The optimal storage path is generated by the path; in the prediction and analysis module, the LSTM time series prediction model is used to model the historical cargo storage information of the target warehouse. The LSTM model contains three hidden layers, and the number of neurons in each layer is 128, 64, and 32, respectively. The input features of the model include the quantity of goods, storage location distribution, turnover rate, and inventory priority, and the output features are the predicted quantity of goods and their storage priority. In the task allocation module, task allocation is performed on the automated equipment in the target warehouse. The automated equipment is matched with the tasks to be executed by constructing a bipartite graph model. The nodes of the bipartite graph model are divided into device nodes and task nodes. The weight of the edge represents the cost of the equipment executing the task. The equipment scheduling plan is generated by finding the maximum weight match.

[0010] In summary, the present invention realizes the full-process intelligent management of industrial stereoscopic warehouses by combining multidimensional data processing, path planning algorithms, priority allocation strategies and time series prediction models, optimizes resource scheduling and task allocation, improves the system's real-time response capabilities and multi-task concurrent processing capabilities, and meets the modern industry's demand for efficient and intelligent warehousing management. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a structural block diagram of the AI ​​intelligent management system for industrial warehouses of the present invention.

[0012] Figure 2 Schematic diagram of the workflow of the path optimization module of the present invention.

[0013] Figure 3 This is a structural diagram of the LSTM time series prediction model of the prediction analysis module of the present invention.

[0014] Figure 4 This is a task scheduling diagram of the task allocation module of the present invention.

[0015] The accompanying drawings are numbered as follows:

[0016] 1. Data acquisition module; 2. Parameter configuration module; 3. Path optimization module; 4. Prediction and analysis module; 5. Task allocation module; 6. Distributed sensor network; 7. Two-dimensional grid map; 8. LSTM model; 9. Bipartite graph model; 10. Automation equipment. DETAILED DESCRIPTION

[0017] The present invention provides an AI intelligent management method and system for industrial warehouses, and its specific implementation is described in detail with reference to the accompanying drawings. Figure 1 As shown, the system includes a data acquisition module 1, a parameter configuration module 2, a path optimization module 3, a prediction and analysis module 4, and a task allocation module 5. These modules jointly realize the intelligent management of the industrial warehouse through logical connection and data interaction.

[0018] The data acquisition module 1 completes the real-time collection of cargo storage information in the target warehouse through the distributed sensor network 6. The distributed sensor network 6 is composed of multiple pressure sensors, temperature sensors and RFID tag readers installed on the shelves. The pressure sensor is used to detect the weight change of the cargo, the temperature sensor is used to monitor the storage environment conditions of the cargo, and the RFID tag reader records the unique identification information of the cargo. These sensors are distributed in different areas of the warehouse according to a preset spatial layout and upload the collected data to the data acquisition module 1. The data acquisition module 1 is responsible for sorting and outputting the cargo storage information corresponding to each historical time interval before the current time interval. Each piece of information contains the quantity, storage location distribution, turnover rate and inventory priority of each type of cargo in the corresponding historical time interval. It should be noted that the current time interval and the previous historical time interval together constitute a complete time segment, and the duration of each time interval is equal.

[0019] Parameter Configuration Module 2 is used to obtain various operating parameters of the target warehouse, including the warehouse's maximum storage capacity, number of shelves, number of automated equipment, and inlet and outlet throughput. By interfacing with the warehouse management system, Parameter Configuration Module 2 updates and stores these operating parameters in real time. These parameters provide the foundation for subsequent route optimization and task scheduling. A data transmission relationship exists between Parameter Configuration Module 2, Route Optimization Module 3, and Prediction Analysis Module 4, ensuring timely access to various operating parameters.

[0020] The path optimization module 3 dynamically optimizes the storage location of goods in the target warehouse and generates the optimal storage path. Figure 2As shown, the path optimization module 3 first constructs a two-dimensional grid map 7 based on the spatial layout of the target warehouse, where each grid cell corresponds to a storage area within the warehouse. The optimal storage path is generated by calculating the minimum cost path from the starting point to the end point. The path cost function comprehensively considers factors such as distance, storage priority, and the number of shelf levels. The path optimization module 3 transmits the generated optimal storage path as an output to the prediction analysis module 4 for subsequent task allocation.

[0021] The prediction analysis module 4 uses the LSTM time series prediction model 8 to model the historical cargo storage information of the target warehouse, thereby predicting the quantity of various types of cargo received by the target warehouse in the current time interval and their storage priority. Figure 3 As shown, the LSTM model 8 includes three hidden layers, with 128, 64, and 32 neurons in each layer, respectively. The model's input features include cargo quantity, storage location distribution, turnover rate, and inventory priority, and the output features are the predicted cargo quantity and storage priority. In each training run of the LSTM model 8, the known cargo quantity and storage priority of each type received by the target warehouse in a certain past time interval are used as the model's output content, and the cargo storage information corresponding to each historical time interval before the past time interval, the target warehouse's various operating parameters, and the duration of the time interval are used as the model's input content to complete this training action. The prediction analysis module 4 passes the prediction results to the task allocation module 5 to provide a basis for task allocation for the automated equipment.

[0022] The task allocation module 5 performs task allocation on the automation equipment 10 in the target warehouse and generates an equipment scheduling plan. Figure 4 As shown, a bipartite graph model 9 is constructed to match automated devices 10 with tasks to be executed. The nodes in bipartite graph model 9 are divided into device nodes and task nodes, and the edge weights represent the cost of the device executing the task. By finding the maximum weighted match, a device scheduling plan is generated, ensuring efficient utilization of limited resources. The task allocation module 5, based on the prediction results provided by the prediction analysis module 4 and the optimal storage path generated by the path optimization module 3, ultimately develops a detailed device scheduling plan.

[0023] In practice, the system operates as follows: First, the data acquisition module 1 collects the target warehouse's cargo storage information in real time through the distributed sensor network 6, organizes the data, and transmits it to the prediction and analysis module 4. Simultaneously, the parameter configuration module 2 obtains the target warehouse's operating parameters and transmits them to the path optimization module 3 and the prediction and analysis module 4. The path optimization module 3 generates the optimal storage path and passes the result to the prediction and analysis module 4. The prediction and analysis module 4 uses the LSTM model 8 to model historical data, predict the cargo quantity and storage priority for the current time interval, and passes the prediction result to the task allocation module 5. The task allocation module 5 assigns tasks to the automated equipment 10 and ultimately generates an equipment scheduling plan. This entire process implements intelligent management of the entire industrial warehouse process, optimizing resource scheduling and task allocation.

[0024] In this embodiment, all modules are connected via data interfaces, ensuring efficient data transmission between them. For example, the data acquisition module 1 and the prediction and analysis module 4 are connected via a data bus, while the path optimization module 3 and the prediction and analysis module 4 exchange data via a dedicated communication protocol. Furthermore, the specific functionality of each module relies on software programs, which run on high-performance servers to ensure the system's real-time responsiveness and multi-task concurrent processing capabilities.

[0025] This embodiment further refines the specific implementation details of each module. For example, in the data acquisition module 1, the layout of the distributed sensor network 6 must follow certain rules. The installation position of the sensor should cover all storage areas of the warehouse, and the distance between adjacent sensors should meet the signal coverage requirements. In the path optimization module 3, the construction of the two-dimensional grid map 7 must be based on the actual spatial layout of the warehouse, and the size of each grid unit should match the actual storage area of ​​the shelf. In the prediction analysis module 4, the training process of the LSTM model 8 must use a large-scale historical data set to improve the prediction accuracy of the model. In the task allocation module 5, the construction of the bipartite graph model 9 must comprehensively consider the performance parameters of the automation equipment 10 and the complexity of the task to ensure the rationality of task allocation.

[0026] This implementation method realizes the full-process intelligent management of industrial warehouses by combining multi-dimensional data processing, path planning algorithms, priority allocation strategies and time series prediction models, optimizes resource scheduling and task allocation, improves the system's real-time response capabilities and multi-task concurrent processing capabilities, and meets the modern industry's demand for efficient and intelligent warehouse management.

[0027] In order to better enable relevant personnel in this technical field to fully understand and implement the present invention, the specific implementation principle of the present invention is further supplemented below with reference to a specific application scenario.

[0028] In a large industrial warehouse, there are many types of goods and their storage locations are scattered. Warehouse management faces complex requirements such as dynamic inventory optimization, equipment coordination and scheduling, and multi-dimensional data analysis. The warehouse uses the AI ​​intelligent management system of the present invention to carry out full-process intelligent management. The specific operation process is as follows:

[0029] First, data acquisition module 1 collects real-time cargo storage information from the target warehouse through distributed sensor network 6. Pressure sensors in distributed sensor network 6 detect weight changes on shelves, temperature sensors monitor storage conditions, and RFID tag readers record the unique identification information of the goods. These sensors are distributed across different areas of the warehouse according to a pre-set spatial layout and upload the collected data to data acquisition module 1. Data acquisition module 1 compiles and outputs cargo storage information corresponding to each historical time interval prior to the current time interval. Each piece of information includes the quantity, storage location distribution, turnover rate, and inventory priority of each type of goods within the corresponding historical time interval. For example, within the previous hour, the system recorded that Class A goods were stored on shelf level 3 and Class B goods on shelf level 5. Based on their inbound and outbound frequencies, the system then calculates their respective turnover rates and priorities. This fundamental data provides comprehensive support for subsequent route optimization and task allocation.

[0030] At the same time, parameter configuration module 2 acquires various operating parameters of the target warehouse, including the warehouse's maximum storage capacity, number of racking levels, number of automated equipment, and inlet and outlet throughput. These parameters are updated in real time through integration with the warehouse management system and transmitted to route optimization module 3 and prediction analysis module 4. For example, the system determines that the warehouse's maximum storage capacity is 1,000 tons, the racking is divided into 10 levels, the number of automated equipment is 20, and the inlet and outlet throughput is 50 tons per hour. These parameters provide the necessary constraints for subsequent route planning and task scheduling.

[0031] The path optimization module 3 dynamically optimizes the storage location of goods in the target warehouse and generates the optimal storage path. Figure 2 As shown, the path optimization module 3 first constructs a two-dimensional grid map 7 based on the spatial layout of the target warehouse, with each grid cell corresponding to a storage area within the warehouse. By comprehensively considering factors such as distance, storage priority, and the number of shelf levels, it calculates the minimum cost path from the starting point to the end point. For example, when a batch of high-priority Class C goods needs to be transported from the warehouse entrance to the seventh-level shelf, the system calculates that the optimal path is from the warehouse entrance through the fourth-level shelf to the seventh-level shelf, avoiding congestion caused by low-priority goods occupying the aisle. The generated optimal storage path is passed as an output result to the prediction analysis module 4 for subsequent task allocation.

[0032] The prediction analysis module 4 uses the LSTM time series prediction model 8 to model the historical cargo storage information of the target warehouse, thereby predicting the quantity of various types of cargo received by the target warehouse in the current time interval and their storage priority. Figure 3 As shown, LSTM model 8 contains three hidden layers, with 128, 64, and 32 neurons in each layer, respectively. The model's input features include cargo quantity, storage location distribution, turnover rate, and inventory priority, while its output features are the predicted cargo quantity and storage priority. For example, based on historical data from the past 24 hours, the system predicts that Class A cargo will increase by 10 tons within the current time interval, with a high storage priority, while Class B cargo will decrease by 5 tons, with a medium storage priority. During each training run of LSTM model 8, the known cargo quantity and storage priority of each type received by the target warehouse during a certain past time interval are used as the model's output. The model also uses the cargo storage information corresponding to each historical time interval prior to that time interval, the target warehouse's operating parameters, and the duration of each time interval as input to complete the training. The prediction analysis module 4 transmits the prediction results to the task allocation module 5, which provides a basis for task allocation for the automated equipment.

[0033] The task allocation module 5 performs task allocation on the automation equipment 10 in the target warehouse and generates an equipment scheduling plan. Figure 4 As shown, the automated equipment 10 is matched with the tasks to be performed by constructing a bipartite graph model 9. The nodes of the bipartite graph model 9 are divided into equipment nodes and task nodes, and the weight of the edge represents the cost of the equipment to perform the task. For example, when it is necessary to transport Class A goods from the entrance to the 7th shelf, the system calculates that the cost of equipment D1 to perform the task is the lowest based on the equipment performance parameters and the complexity of the task. By finding the maximum weight match, an equipment scheduling plan is generated to ensure the efficient use of limited resources. The task allocation module 5, based on the prediction results provided by the prediction analysis module 4 and the optimal storage path generated by the path optimization module 3, finally formulates a detailed equipment scheduling plan. For example, the system arranges equipment D1 to be responsible for the warehousing task of Class A goods, and equipment D2 to be responsible for the outbound task of Class B goods, and guides the equipment to complete the task quickly through the optimal path provided by the path optimization module 3.

[0034] Throughout the entire process, all modules are connected via data interfaces, ensuring efficient data transmission between them. For example, data acquisition module 1 and prediction and analysis module 4 are connected via a data bus, while route optimization module 3 and prediction and analysis module 4 exchange data via a dedicated communication protocol. Furthermore, the specific functionality of each module relies on software programs, which run on high-performance servers to ensure the system's real-time responsiveness and multi-tasking capabilities.

[0035] Through the above steps, this system achieves intelligent management of the entire industrial warehouse process. For example, under high-load scenarios, the system can dynamically adjust storage paths and equipment scheduling based on real-time inventory information and predicted future demand, avoiding inefficiencies caused by irrational resource allocation. Furthermore, by combining multidimensional data processing, path planning algorithms, priority allocation strategies, and time series prediction models, the system significantly improves the efficiency of resource scheduling and task allocation, meeting the modern industrial demand for efficient and intelligent warehouse management.

[0036] In summary, the present invention solves the problems existing in traditional warehouse management, such as insufficient dynamic inventory optimization, low efficiency of equipment collaborative scheduling, and limited multi-dimensional data analysis capabilities, through specific operation steps and algorithm implementation, and provides reliable technical support for the intelligent management of industrial warehouses.

[0037] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0038] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An AI intelligent management method for industrial warehouses, characterized in that: The method comprises the following steps: Collecting the target warehouse's cargo storage information corresponding to each historical time interval before the current time interval, where the current time interval starts at the current moment. The cargo storage information corresponding to each historical time interval includes the quantity, storage location distribution, turnover rate, and inventory priority of each type of cargo received by the target warehouse during the historical time interval. Obtaining various operating parameters of the target warehouse, including the warehouse's maximum storage capacity, number of shelves, number of automated equipment, and inlet and outlet throughput; Dynamically optimize the storage location of goods in the target warehouse to generate the optimal storage path, and use the optimal storage path as the output result of the path planning module; The LSTM time series prediction model is used to predict the quantity and storage priority of each type of cargo received by the target warehouse in the current time interval based on the cargo storage information corresponding to each historical time interval before the current time interval, the target warehouse's operating parameters, and the duration of the time interval; Based on the predicted quantity of various types of goods and their storage priorities in the current time interval, tasks are assigned to the automated equipment in the target warehouse and an equipment scheduling plan is generated.

2. The AI ​​intelligent management method for industrial warehouses according to claim 1 is characterized by: The LSTM time series prediction model is used to predict the quantity of each type of goods received by the target warehouse in the current time interval and the storage priority thereof based on the storage information of each piece of goods corresponding to each historical time interval before the current time interval, the operating parameters of the target warehouse, and the duration of the time interval. The method includes: using the known quantity of each type of goods received by the target warehouse in a past time interval and the storage priority thereof as the output content of the LSTM time series prediction model, and using the storage information of each piece of goods corresponding to each historical time interval before the past time interval, the operating parameters of the target warehouse, and the duration of the time interval as the input content of the LSTM time series prediction model to complete this training action.

3. The AI ​​intelligent management method for industrial warehouses according to claim 1 is characterized by: Based on the predicted quantity of various types of goods and their storage priorities in the current time interval, tasks are assigned to the automated equipment in the target warehouse to generate an equipment scheduling plan. This includes: constructing a bipartite graph model, in which the nodes of the bipartite graph model are divided into equipment nodes and task nodes, and the weight of the edge represents the cost of the equipment executing the task. The equipment scheduling plan is generated by finding the maximum weight match.

4. The AI ​​intelligent management method for industrial high-bay warehouses according to claim 1 is characterized by: Collecting each piece of cargo storage information corresponding to each historical time interval before the current time interval in the target warehouse includes: using a distributed sensor network to collect the cargo storage information in the target warehouse in real time, the distributed sensor network is composed of a plurality of pressure sensors, temperature sensors and RFID tag readers installed on the shelves, the pressure sensors are used to detect weight changes of the cargo, the temperature sensors are used to monitor the storage environment conditions of the cargo, and the RFID tag readers are used to record unique identification information of the cargo.

5. The AI ​​intelligent management method for industrial warehouses according to claim 3 is characterized by: The LSTM time series prediction model includes three hidden layers, with 128, 64, and 32 neurons in each layer, respectively. The input features of the model include the quantity of goods, storage location distribution, turnover rate, and inventory priority, and the output features are the predicted quantity of goods and their storage priority.

6. An AI intelligent management system for industrial warehouses, characterized by: The system includes the following components: The data collection module is used to collect the storage information of each cargo in the target warehouse corresponding to each historical time interval before the current time interval; Parameter configuration module, used to obtain various operating parameters of the target warehouse; A path optimization module, connected to the data acquisition module, is used to dynamically optimize the storage location of goods in the target warehouse and generate an optimal storage path; a prediction and analysis module, connected to the data acquisition module, the parameter configuration module, and the path optimization module, respectively, for predicting the quantity and storage priority of each type of cargo received by the target warehouse in the current time interval based on the cargo storage information corresponding to each historical time interval before the current time interval, the operating parameters of the target warehouse, and the duration of the time interval using an LSTM time series prediction model; A task allocation module, connected to the forecast analysis module, is used to perform task allocation on the automated equipment in the target warehouse based on the predicted quantity of each type of goods and their storage priority in the current time interval, and generate an equipment scheduling plan; The collection of each piece of cargo storage information corresponding to each historical time interval before the current time interval of the target warehouse further includes: each historical time interval before the current time interval and the current time interval together form a complete time segment, and the duration of each time interval is equal.

7. The AI ​​intelligent management system for industrial warehouses according to claim 6 is characterized by: The data acquisition module completes real-time collection of cargo storage information in the target warehouse through a distributed sensor network. The distributed sensor network consists of multiple pressure sensors, temperature sensors, and RFID tag readers installed on the shelves. The pressure sensors are used to detect weight changes of the cargo, the temperature sensors are used to monitor the storage environment conditions of the cargo, and the RFID tag readers are used to record the unique identification information of the cargo.

8. The AI ​​intelligent management system for industrial warehouses according to claim 6 is characterized by: The path optimization module constructs a two-dimensional grid map based on the spatial layout diagram of the target warehouse. Each grid cell in the two-dimensional grid map corresponds to a storage area in the warehouse, and generates an optimal storage path by calculating the minimum cost path from the starting point to the end point.

9. The AI ​​intelligent management system for industrial warehouses according to claim 6 is characterized by: The prediction and analysis module uses an LSTM time series prediction model to model the historical cargo storage information of the target warehouse. The LSTM model contains three hidden layers, with 128, 64, and 32 neurons in each layer, respectively. The input features of the model include cargo quantity, storage location distribution, turnover rate, and inventory priority, and the output features are the predicted cargo quantity and its storage priority.

Images