Intelligent warehouse management method and system based on Internet of Things and artificial intelligence

Abstract

An intelligent warehouse management method based on the Internet of Things and artificial intelligence comprises the steps that 1) intelligent inventory management is carried out, and each commodity which is put in storage and put out of storage is automatically identified and tracked by utilizing a cargo container label and an image identification technology; the artificial intelligence technology is combined with big data analysis, future inventory demands are predicted, the inventory level is automatically adjusted, and the inventory cost and the stockout risk are reduced; 2) an automatic operation process: monitoring the equipment state in real time through the Internet of Things technology, ensuring the efficient operation of the equipment, and reducing the downtime; 3) performing real-time data analysis and decision support, and collecting various data generated in the storage process, including inventory data, order data and equipment operation data; deep learning and analysis are performed on storage data by using an artificial intelligence technology, intelligent decisions such as inventory prediction and path planning are realized, and storage management and logistics distribution processes are optimized; and 4) cross-system collaboration and information interconnection of a plurality of systems such as data sharing, storage, transportation, sales and the like are carried out to form end-to-end supply chain collaboration.

Description

Technical Field

[0001] The present invention relates to the field of warehouse management technology, and in particular to a method and system for intelligent warehouse management based on the Internet of Things and artificial intelligence. Background Art

[0002] (1) Causes of the technology of the present invention:

[0003] With the rapid development of e-commerce and logistics industries, the demand for warehouse management is growing. Traditional warehouse management systems mainly rely on manual operations, which are inefficient and prone to errors. In addition, traditional systems have many deficiencies in information integration, real-time monitoring and intelligent management. In order to meet these challenges and improve the efficiency and accuracy of warehouse management, the present invention comes into being.

[0004] (2) The present invention mainly solves the following problems:

[0005] High dependence on manual labor: Traditional warehousing systems are highly dependent on manual operations, resulting in low efficiency and prone to human errors.

[0006] Information islands: The lack of information communication between subsystems leads to chaotic data management and difficulty in achieving overall optimization.

[0007] Lack of real-time monitoring: It is impossible to monitor inventory and logistics status in real time, making it difficult to adjust management strategies in a timely manner.

[0008] (3) Among the existing public technologies, there are also some warehouse management systems based on the Internet of Things and artificial intelligence, but they have the following shortcomings and disadvantages:

[0009] Limited data processing capabilities: Many existing systems have limitations in data processing capabilities and cannot process large-scale IoT data in real time.

[0010] Low level of intelligence: Although the existing system has introduced some intelligent technologies, its decision support and scheduling optimization capabilities still need to be improved.

[0011] Low system integration: Existing systems often require multiple independent subsystems to work together, with low integration and complex management.

[0012] Poor user experience: The existing management interface is not user-friendly, the operation is complicated, and it is not convenient for managers to monitor and operate in real time.

[0013] High cost: The implementation of some advanced functions requires high hardware and software investment, which increases the operating costs of the enterprise.

[0014] In summary, the present invention provides an efficient, intelligent and easy-to-use warehouse management system by integrating the Internet of Things and artificial intelligence technologies, which solves many deficiencies in the prior art. Summary of the invention

[0015] In order to solve the problems existing in the prior art, the purpose of the present invention is to realize intelligent management of overseas warehouses by integrating advanced Internet of Things (IoT) technology, artificial intelligence technology (AI) and data intelligent analysis technology, so as to improve warehousing efficiency, reduce costs and improve customer service level.

[0016] The technical solution of the present invention is a method of intelligent warehouse management based on the Internet of Things and artificial intelligence, and the specific implementation steps are as follows:

[0017] Step 1: Smart Inventory Management

[0018] Utilize cargo box labels and image recognition technology to automatically identify and track every item entering and leaving the warehouse.

[0019] Use artificial intelligence technology combined with big data analysis to predict future inventory demand, automatically adjust inventory levels, and reduce inventory costs and out-of-stock risks.

[0020] First, obtain real-time inventory, inbound and outbound orders, in-transit inventory, and supplier sales data; second, transform the original data and obtain data and trends with different characteristics based on natural language processing, principal component analysis and other technologies; finally, use time series analysis to analyze the changes in past inventory demand at different time points to predict basic demand, and use linear regression models with different characteristics to predict the corresponding inventory demand.

[0021] Step 2: Automated workflow

[0022] Introduce automated picking, conveying and other systems to reduce labor costs, improve warehouse operation efficiency and accuracy, and reduce human errors and losses.

[0023] Use IoT technology to monitor equipment status in real time to ensure efficient operation of equipment and reduce downtime.

[0024] Step 3: Real-time data analysis and decision support

[0025] Collect various types of data generated during the warehousing process, including inventory data, order data, equipment operation data, etc.

[0026] Use artificial intelligence technology to conduct deep learning and analysis of warehouse data, realize intelligent decision-making such as inventory forecasting and route planning, and optimize warehouse management and logistics distribution processes.

[0027] A complete directed graph model is constructed based on the actual layout of the warehouse and equipment parameters, where nodes represent storage locations, shelves, entrances and exits, packing tables, etc., and edges represent channels between nodes, with weights assigned to each edge (such as distance, travel time, handling cost, etc.). When the warehouse layout is adjusted or equipment performance changes, the nodes, edges, and weights in the graph are adjusted in a timely manner. When the system generates a delivery task, the system uses a path search algorithm to plan the optimal path for the equipment based on the graph theory model.

[0028] Step 4: Cross-system collaboration and data sharing

[0029] Connect multiple systems such as warehousing, transportation, and sales to form end-to-end supply chain collaboration. Based on cloud computing, big data and other technologies, realize cross-regional and cross-system data sharing and business collaboration.

[0030] Beneficial effects:

[0031] (1) The present invention uses intelligent inventory management, cargo box labels and image recognition technology to achieve accurate tracking and management of each item. Combined with big data analysis, it can accurately predict future inventory needs and automatically adjust inventory levels, thereby effectively reducing inventory costs and out-of-stock risks;

[0032] (2) The present invention significantly reduces labor costs and improves the efficiency and accuracy of warehousing operations by introducing automated picking and conveying systems. The function of real-time monitoring of equipment status ensures efficient operation of the equipment, reduces downtime due to failures, and further improves overall operational efficiency;

[0033] (3) The present invention collects and analyzes various types of data generated during the warehousing process through real-time data analysis and decision support, and uses artificial intelligence technology for deep learning and analysis to achieve intelligent decisions such as inventory forecasting and route planning (for example, time series analysis, by analyzing past inventory change trends, recommending time series models, combining seasonal decomposition methods, etc., to predict future inventory demand). This not only optimizes the warehousing management and logistics distribution process, but also provides a scientific decision-making basis for management, improving the accuracy and timeliness of decision-making;

[0034] (4) Through cross-system collaboration and data sharing, the present invention has opened up the information interconnection of multiple systems such as warehousing, transportation, and sales, forming an end-to-end supply chain collaboration. Based on cloud computing, big data and other technologies, cross-regional and cross-system data sharing and business collaboration are realized, which enhances the transparency and collaborative efficiency of the entire supply chain and improves the market competitiveness of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a system architecture diagram for implementing an intelligent warehouse management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention;

[0036] Figure 2 This is a flowchart of an automated operation implemented by an intelligent warehouse management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention;

[0037] Figure 3 A flow chart of cross-system collaboration and data sharing implemented by an intelligent warehouse management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention;

[0038] Figure 4 The present invention is a flowchart of an implementation method of an optimal path planning model implemented by an intelligent warehouse management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] Figure 1 This is a system architecture diagram for implementing an intelligent warehousing management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention. The system architecture includes the following modules.

[0040] (1) OMS seller side: mainly used to handle order management. It allows sellers to receive, process and manage orders from different channels, ensuring that orders can be accurately delivered to the warehouse for processing. The main functions include order reception and processing, automatic inventory checking, customer service and data analysis;

[0041] (2) OSS management: mainly for internal operation management personnel, providing a series of support tools and services to optimize business processes and improve work efficiency. The main functions include operation monitoring, cost control, process optimization, operation report generation, etc.

[0042] (3) WMS warehouse: focuses on the operation management within the warehouse, aiming to improve the efficiency and accuracy of warehouse management through advanced technical means. The main functions include storage location management, inbound management, outbound management, inventory management, automated picking, inventory management, etc.

[0043] Figure 2 This is a flowchart of the automated operation implemented by the intelligent warehousing management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention, and the flowchart includes two processes: warehousing in and out.

[0044] Automated operation process: Product and storage location information is already available during warehousing and outbound delivery, but each warehousing order or outbound delivery order has multiple products, and each product will be placed in multiple storage locations. Therefore, both the warehousing and shelving tasks and the outbound picking tasks will involve a complete document operation sequence (products and storage locations). The directed graph model finds the optimal path, which is to find the optimal operation path among various combinations of products and storage locations.

[0045] Warehouse entry process: Before the seller delivers the goods to the warehouse, he / she submits a warehouse entry application through the system. After the goods arrive at the warehouse, the warehouse staff uses an RF handheld terminal to scan and read the cargo box or product label of the goods to verify the cargo information. Through the system-generated shelf tasks, combined with the directed graph optimal path algorithm, the shelf personnel are prompted to store the goods in the designated warehouse location according to the optimal route and sequence. At the same time, the system automatically updates the inventory information and notifies the seller;

[0046] Outbound process: After receiving the seller's order information, the system automatically checks the inventory status and generates picking tasks and optimal picking route suggestions in combination with the directed graph optimal path algorithm. The pickers use the RF handheld terminal to pick the goods according to the optimal path. After the system automatically obtains the last-mile delivery note, the staff scans the package information to confirm and then loads the goods for shipment. After the shipment is completed, the system automatically updates the inventory and notifies the seller.

[0047] Figure 3 This is a flow chart of cross-system collaboration and data sharing implemented by the intelligent warehousing management method based on the Internet of Things and artificial intelligence in an embodiment of the present invention, and the flow chart includes the following parts.

[0048] (1) Data collection layer: IoT devices (such as handheld RF guns, cargo tags, etc.) collect real-time data in the warehouse, including inventory status, operation status, etc.; the sales order channel receives order information from the sales platform or the seller's ERP system;

[0049] (2) Data processing layer: data cleaning and preprocessing to ensure data quality and availability; use AI algorithms for data analysis, such as predicting demand and optimizing inventory; automatically select the best last-mile logistics service provider and generate express delivery bills; collect and analyze various types of data generated during the warehousing process, use artificial intelligence technology for deep learning and analysis, and realize intelligent decision-making such as inventory forecasting and route planning. Through time series analysis, by analyzing past inventory change trends, a time series model is recommended, combined with seasonal decomposition method, to predict future inventory demand;

[0050] (3) Business logic layer: After receiving sales order information, the system automatically triggers operations such as picking and packing; adjusts inventory levels based on analysis results to ensure reasonable inventory; integrates last-mile logistics channel resources, optimizes delivery routes, and provides delivery progress query services;

[0051] (4) Data sharing and interaction layer: Build unified data standards and protocols to ensure information exchange between different systems; use middleware technology as a bridge between modules to promote efficient collaboration between heterogeneous systems; set up permission control mechanisms to ensure the safe use of data;

[0052] (5) User interface layer: Provides a user-friendly interface so that managers can easily view information such as warehouse status and order processing progress, making it convenient to monitor and manage at any time.

[0053] A complete directed graph model is constructed based on the actual layout and equipment parameters of the warehouse, in which nodes represent storage locations, shelves, entrances and exits, packaging tables, etc., and edges represent channels between nodes. Weights are assigned to each edge (such as distance, travel time, transportation cost, etc.).

[0054] Figure 4 The flowchart of the implementation method of the optimal path planning model implemented by the intelligent warehouse management method based on the Internet of Things and artificial intelligence in the embodiment of the present invention is divided into the following steps.

[0055] Step 1: Initialization

[0056] First, the directed graph abstracted from the warehouse layout needs to be represented in a suitable way, usually in the form of an adjacency matrix. Assuming that there are n nodes in the directed graph corresponding to the warehouse layout (representing different storage locations, facilities including storage locations, shelves, entrances and exits, packing tables, etc.), create an n*n two-dimensional array to represent the adjacency matrix of the graph. Create an n*n two-dimensional array to represent the distance matrix, which is used to record the estimated value of the shortest path length between any two nodes. Create another n*n two-dimensional array as a predecessor matrix to record the predecessor node of each node in the shortest path;

[0057] Step 2: Iterate to find the shortest path

[0058] The shortest distance matrix between any two points in the graph is gradually updated through a triple loop. The outer loop (intermediate node traversal) checks in turn whether each node can shorten the path length between other node pairs when used as an intermediate node. The middle loop (starting node traversal) traverses all possible starting nodes to check whether the current intermediate node can shorten the path from the starting node to other nodes. The inner loop (target node traversal) checks whether the path length from the starting node to the target node can be shortened by passing the current intermediate node. When each three-layer nested loop reaches the inner layer, it determines whether the path length from the starting node to the target node can be shortened by passing the intermediate node. If so, the result is updated;

[0059] Step 3: Repeat until the target node is found

[0060] Continue the three-layer nested loop in step 2, traversing all possible values ​​in turn until all node combinations have been checked, that is, the update process of the entire distance matrix and predecessor matrix is ​​completed. This process will consider all nodes as intermediate nodes, so as to comprehensively find the shortest path length between any two nodes after all possible intermediate nodes are optimized and the corresponding predecessor node information.

[0061] Step 4: Construct the shortest path

[0062] When there is a specific path planning requirement, such as picking up goods or transporting goods from a certain starting location in the warehouse to a certain target location, first identify the starting and ending nodes, start from the ending node to query the direct predecessor node on the shortest path, treat this predecessor node as the ending node, and use the same method to obtain the predecessor nodes in sequence, and finally reverse the sequence to obtain the shortest path.

[0063] The characteristics of the adjacency list of a directed graph are as follows. If a directed graph has n nodes and e edges, then the node table has n nodes and the adjacency table has e nodes. The out-degree of a node is the number of nodes in the singly linked list behind the node.

[0064] Figure 4 In the embodiment, a flow chart of a method for implementing an optimal path planning model implemented by an intelligent warehouse management method based on the Internet of Things and artificial intelligence is provided.

[0065] Example description: In the order picking mode, the optimal picking path is planned based on a delivery order (this delivery order contains multiple goods, and each kind of goods is stored in multiple different storage locations).

[0066] 1. Warehouse layout setting

[0067] 1) The starting point and end point of the optimal path for picking outbound orders: the starting point S is the fixed picking start position or the current real-time position of the picker, and the end point E is the packaging review area (picking temporary storage area);

[0068] 2) Possible intermediate nodes of the optimal picking path: For example, product A has 3 storage locations, namely A1, A2, and A3; product B has 2 storage locations, namely B1 and B2; product C has 2 storage locations, namely C1 and C2;

[0069] 3) The weight of each minimum path in the optimal picking path: The channels between nodes in the warehouse (including the starting point, the end point, and the intermediate nodes) are regarded as edges, and the weight of the edge is set as the walking time between two nodes, and a directed graph is constructed. Assume that the weights between the nodes are as follows (partial examples):

[0070]

[0071]

[0072] 2. Initialization

[0073] Set the distance array, record the shortest distance from the starting point S to each node and set the initial value. Set the visited node set, record the distance array of all possible paths;

[0074] 3. Iterative planning of the optimal path

[0075] Select the node closest to the starting point S from the unvisited nodes (for example, A1 has a distance weight of 3) and add it to the set of visited nodes. Then select the node closest to A1 from the unvisited nodes (for example, B1 has a distance weight of 5), so the distance weight of S-A1-B1 is 8. Continue to select the node closest to B1 from the unvisited nodes (for example, C2 has a distance weight of 4), so the distance weight of S-A1-B1-C2-E is 14.

[0076] Continue to repeat the above steps, select the node closest to S and the node closest to A and B from the unvisited nodes in turn, and get a complete path distance weight.

[0077] When all nodes have been visited, the shortest distance and optimal path from the starting point S to each product location and then to the end point E can be obtained from the distance array.

[0078] 4. Final Result

[0079] Assume that the optimal path is S-A1-B1-C2-E, which means taking product A from warehouse A1, product B from warehouse B1, and product C from warehouse C2 in this order, and finally sending all the products to packing and review area E. This is the optimal operation path for picking the outbound order, and the total walking time taken by this path is the shortest.

[0080] Although the present invention has been described above, those skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A method for intelligent warehouse management based on the Internet of Things and artificial intelligence, characterized in that the steps as follows: Step 1: Smart Inventory Management Automatically identify and track every item entering and leaving the warehouse using cargo box labels and image recognition technology; Use artificial intelligence technology combined with big data analysis to predict future inventory needs, automatically adjust inventory levels, and reduce inventory costs and out-of-stock risks; First, obtain real-time inventory, inbound and outbound orders, in-transit inventory, and supplier sales data; Secondly, the original data is transformed, and data and trends with different characteristics are obtained based on natural language processing, principal component analysis and other technologies. Finally, the time series analysis method is used to analyze the changing patterns of past inventory demand at different time points to predict basic demand, and the corresponding inventory demand is predicted through linear regression models with different characteristics. Step 2: Automated workflow Introducing automated picking and conveying systems to reduce labor costs, improve warehouse operation efficiency and accuracy, and reduce human errors and losses; Use IoT technology to monitor equipment status in real time to ensure efficient operation of equipment and reduce downtime; Step 3: Real-time data analysis and decision support Collect various data generated during the warehousing process, including inventory data, order data, equipment operation data, etc.; Use artificial intelligence technology to conduct deep learning and analysis of warehouse data, realize intelligent decision-making such as inventory forecasting and route planning, and optimize warehouse management and logistics distribution processes; A complete directed graph model is constructed based on the actual layout of the warehouse and equipment parameters, where nodes represent storage locations, shelves, entrances and exits, and packaging tables, and edges represent channels between nodes, with weights assigned to each edge. When the warehouse layout is adjusted or equipment performance changes, the nodes, edges, and weights in the graph are adjusted in a timely manner. When the system generates a delivery task, the system uses a path search algorithm based on the graph theory model to plan the optimal path for the equipment. Step 4: Cross-system collaboration and data sharing The information interconnection of multiple systems such as warehousing, transportation, and sales forms end-to-end supply chain collaboration.

2. The method of intelligent warehouse management based on the Internet of Things and artificial intelligence according to claim 1 is characterized in that: The product and storage location information is already available during the inbound and outbound operations. However, each inbound or outbound order has multiple products, and each product is placed in multiple storage locations. Therefore, both the inbound shelf task and the outbound picking task involve a complete document operation sequence. The directed graph model searches for the optimal path, which is to find the optimal operation path among various combinations of products and storage locations. Warehouse entry process: Before the seller delivers the goods to the warehouse, he / she submits a warehouse entry application through the system. After the goods arrive at the warehouse, the warehouse staff uses an RF handheld terminal to scan and read the cargo box or product label of the goods to verify the cargo information. Through the system-generated shelf tasks, combined with the directed graph optimal path algorithm, the shelf personnel are prompted to store the goods in the designated warehouse location according to the optimal route and sequence. At the same time, the system automatically updates the inventory information and notifies the seller; Outbound process: After receiving the seller's order information, the system automatically checks the inventory status and generates picking tasks and optimal picking route suggestions in combination with the directed graph optimal path algorithm; the pickers use RF handheld terminals to pick the goods according to the optimal path. After the system automatically obtains the last-mile delivery note, the staff scans the package information to confirm and then loads the goods for shipment. After the shipment is completed, the system automatically updates the inventory and notifies the seller.

3. The method of intelligent warehouse management based on the Internet of Things and artificial intelligence according to claim 1 is characterized in that: In the data processing layer: AI algorithms are used for data analysis, and artificial intelligence technology is used for deep learning and analysis to achieve intelligent decisions such as inventory forecasting and route planning. Through time series analysis, by analyzing past inventory change trends, time series models are recommended, combined with seasonal decomposition method, to predict future inventory demand.

4. A system for performing the steps of the method for intelligent warehousing management based on the Internet of Things and artificial intelligence according to any one of claims 1-3.

Images