Intelligent warehouse management method for electronic products
Through automation and intelligent technology, the warehousing management of electronic products has been optimized, and the problems of low production efficiency and inventory management are solved, efficient storage, accurate sorting and timely delivery are achieved, operating costs are reduced, product quality and customer satisfaction are improved .
Patent Information
- Application Number
- CN202411041690.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-22
AI Technical Summary
The existing electronic product warehousing management has problems such as low production efficiency, difficulty in inventory management, insufficient product quality control, poor order processing, low space utilization and high cost, which affects the company's operating efficiency and cost control.
Adopt automation technology and intelligent systems, combining machine vision systems, RFID technology, intelligent sorting systems and machine learning models, optimize warehousing processes, realize real-time quality inspection, inventory tracking, space utilization and order processing, and reduce manual intervention and operation costs.
Improve production efficiency and product quality, reduce operating costs, enhance inventory visual management, shorten order processing time, and improve supply chain response speed and customer satisfaction.
Smart Images

Figure CN119005859A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of logistics management, and in particular to an intelligent warehousing management method for electronic products. Background Art
[0002] Warehouse Management is the management of warehouses and the materials stored in them, and the storage and preservation of goods through warehouses. Warehouses in modern enterprises have become the logistics centers of enterprises, and are used by enterprises as a bridge connecting suppliers and demanders. They are not only regarded as part of the process of forming added value, but also as a key factor in the successful operation of enterprises. From the perspective of suppliers, warehouses as distribution centers are engaged in efficient distribution processing, inventory management, transportation and distribution. From the perspective of demanders, warehouses as distribution centers must meet the needs of various customers with the greatest flexibility and timeliness. Therefore, for enterprises, warehouse management is of great significance.
[0003] Under the new economic and competitive situation, while focusing on efficiency and constantly exploring and developing their own competitiveness, enterprises have increasingly noticed the importance of reasonable warehouse management. Accurate warehouse management can effectively control and reduce circulation and inventory costs, and is the key to help and guarantee for enterprises to maintain their advantages.
[0004] At present, the existing electronic product warehouse management has problems such as low production efficiency and difficult inventory management, insufficient product quality control, unsmooth order processing and logistics transportation, low space utilization, high cost, and large amount of data that is prone to errors. This increases labor costs and long inventory cycles, affecting the timeliness of delivery and bringing adverse effects to the operation of enterprises. Therefore, it is urgent to study an intelligent warehouse management method to simplify the manual operation process and improve the production efficiency of enterprises. Summary of the invention
[0005] The object of the present invention is to provide an intelligent warehousing management method for electronic products with high production efficiency, low operating cost, and the ability to ensure efficient storage, accurate sorting, and timely delivery of products.
[0006] The technical solution to achieve the purpose of the present invention is: an intelligent storage management method for electronic products, comprising the following steps:
[0007] Step 1: The manufacturing department produces the product;
[0008] Step 2: The quality inspection system uses a machine vision system to inspect the product and feeds the inspection results back to the production process to improve production efficiency and product quality.
[0009] Step 3: Primary packaging of qualified products;
[0010] Step 4: Store the qualified products in primary packaging in the primary finished product warehouse;
[0011] Step 5: Determine whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and perform secondary sorting on the finished products in the primary finished product warehouse according to the work order;
[0012] Step 6: Secondarily package the secondary sorted products and store them in the finished product warehouse;
[0013] Step 7: Determine whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and sort the finished products in the finished product warehouse for shipment according to the work order;
[0014] Step 8: Pack the sorted finished products on pallets;
[0015] Step 9: Ship the packaged finished products out of the warehouse.
[0016] Compared with the prior art, the present invention has the following significant advantages:
[0017] (1) The use of automation technology and intelligent systems optimizes the warehousing process, reduces manual intervention, and improves the operating efficiency and speed of the production line;
[0018] (2) Reduced labor costs and operating expenses, and lowered enterprise costs by optimizing inventory management and reducing waste losses;
[0019] (3) Using RFID technology to track product location and status in real time improves inventory visualization and management efficiency and reduces inventory deviations and omissions;
[0020] (4) Through intelligent storage optimization algorithms and machine learning models, warehouse space is maximized, improving storage capacity and inventory turnover;
[0021] (5) Using machine vision systems for product quality inspection reduces the risk of defective products entering the market, improving product quality and customer satisfaction;
[0022] (6) Through real-time monitoring and intelligent sorting systems, order processing and product delivery time are shortened, and the responsiveness and flexibility of the supply chain are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 The figure is a flow chart of an intelligent storage management method for electronic products according to the present invention. DETAILED DESCRIPTION
[0024] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0025] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0026] The present invention provides an intelligent storage management method for electronic products, comprising the following steps:
[0027] Step 1: The manufacturing department produces the product;
[0028] Step 2: The quality inspection system uses a machine vision system to inspect the product and feeds the inspection results back to the production process to improve production efficiency and product quality.
[0029] Step 3: Primary packaging of qualified products;
[0030] Step 4: Store the qualified products in primary packaging in the primary finished product warehouse;
[0031] Step 5: Determine whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and perform secondary sorting on the finished products in the primary finished product warehouse according to the work order;
[0032] Step 6: Secondarily package the secondary sorted products and store them in the finished product warehouse;
[0033] Step 7: Determine whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and sort the finished products in the finished product warehouse for shipment according to the work order;
[0034] Step 8: Pack the sorted finished products on pallets;
[0035] Step 9: Ship the packaged finished products out of the warehouse.
[0036] As a specific example, in step 2, the quality inspection system uses a machine vision system to inspect the product and feeds the quality inspection results back to the production process to improve production efficiency and product quality, as follows:
[0037] Step 2.1: According to the order, the manufacturing department sends the new products to the warehouse automatic line;
[0038] Step 2.2: The visual inspection system obtains product images through the industrial camera installed at the front end of the warehouse automatic line;
[0039] Step 2.3, preprocessing the acquired product image;
[0040] Step 2.4: Use image processing and computer vision techniques to extract features from the preprocessed image;
[0041] Step 2.5: Based on the extracted features, use the deep learning model to perform defect detection and classification;
[0042] Step 2.6: Based on the model output results, determine whether the product is qualified and what defects the unqualified products have;
[0043] Step 2.7: The quality inspection system records the quality inspection results of each product and performs data analysis;
[0044] Step 2.8: Feed the results of the quality inspection system back to the production process, and improve the product quality on the production line through automated equipment adjustments and manual intervention.
[0045] As a specific example, in step 2.5, based on the extracted features, a deep learning model is used to perform defect detection and classification as follows:
[0046] Step 2.5.1, collect image datasets containing normal products and different types of defects;
[0047] Step 2.5.2, preprocess the images in the image dataset, including denoising, contrast enhancement and size standardization;
[0048] Step 2.5.3, build a CNN model, including convolutional layer, pooling layer, batch normalization layer and fully connected layer;
[0049] Step 2.5.4: Use the cross entropy loss function and Adam optimizer to train the model and adjust the hyperparameters to optimize the model performance.
[0050] Step 2.5.5, output the classification results of defect types;
[0051] Step 2.5.6: Evaluate the accuracy, precision, and recall metrics of the model;
[0052] Step 2.5.7: Deploy the trained model to the quality inspection system for defect detection and classification, and monitor and continuously optimize the model to adapt to new product changes and defect types.
[0053] As a specific example, in step 2.7, the quality inspection system records the quality inspection results of each product and performs data analysis as follows:
[0054] Step 2.7.1, collect the quality inspection results of the defect-detected products, including structured data and unstructured data, and store them in the central database;
[0055] Step 2.7.2: Perform data cleaning to handle outliers, missing values, or erroneous values in the data, and extract useful features from the quality inspection data, including the frequency of each defect and the product quality inspection pass rate;
[0056] Step 2.7.3: Analyze the data using machine learning models, use the K-value clustering algorithm to group products according to quality inspection features, identify product groups with similar defect patterns, use the isolation forest algorithm to identify abnormal quality inspection results, and use cross-validation techniques to evaluate the performance of the model, and adjust the model hyperparameters to optimize the algorithm's predictive and generalization capabilities;
[0057] Step 2.7.4: Use real-time data stream processing technology to monitor and analyze the quality inspection process in real time to identify and respond to potential problems;
[0058] Step 2.7.5: Use charts to display key indicators and trends of quality inspection data on the intelligent warehousing system interface to help decision makers understand the data and make corresponding management decisions.
[0059] As a specific example, in step 3, the qualified products are primary packaged as follows:
[0060] Step 3.1: The automatic packing system receives qualified products that have passed quality inspection. The robot uses the feature extraction and pattern recognition algorithm SURF to identify the position and direction of the product in the working area for subsequent grasping and placement.
[0061] Step 3.2: Use the inverse kinematics algorithm to calculate the motion parameters of each joint of the manipulator according to the target position and posture, use the A* algorithm to plan the motion path of the manipulator, and use the GJK algorithm to detect possible collisions during the movement of the manipulator in real time;
[0062] Step 3.3, use sensor data to monitor and adjust the gripping force and posture of the robot in real time to adapt to the characteristics of different products, and use model predictive control MPC;
[0063] Step 3.4: Use Kalman filtering to track targets in a dynamic environment, and control the robot to grab the product and move it to the target location based on visual feedback and sensor data;
[0064] Step 3.5: According to different product sizes and box spaces, use dynamic programming to optimize product placement strategies. Use a machine vision system combined with the object detection model YOLO based on the convolutional neural network CNN to detect and identify products in real time to ensure correct product placement.
[0065] Step 3.6: Use the real-time data provided by sensors and vision systems to detect and provide feedback on abnormalities during the packing process;
[0066] Step 3.7: After placing the products, seal the boxes.
[0067] As a specific example, in step 4, the qualified products of primary packaging are stored in the primary finished product warehouse, as follows:
[0068] Step 4.1: Collect historical order data of the product, including model, size, weight, and sales frequency information;
[0069] Step 4.2: Extract product features, including size, weight, classification, sales trend, warehouse temperature and humidity, as input features of the model;
[0070] Step 4.3: Use historical order data to train a deep neural network (DNN) model to predict the optimal storage location and cargo allocation for different products.
[0071] Step 4.4: Deploy the trained deep neural network DNN model to the real-time system to monitor new orders and inventory changes in real time, predict and optimize product location allocation and storage strategies based on real-time data, ensure timely fulfillment of order requirements and optimize warehouse space utilization;
[0072] Step 4.5: Dynamically adjust the parameters and strategies of the machine learning model based on the data feedback from actual operation to optimize the storage efficiency and operation process of the product;
[0073] Step 4.6: Use evaluation indicators to evaluate the effectiveness of the machine learning model, and adjust and optimize the machine learning model based on the evaluation results to improve the product's storage location management strategy to meet changing market demands and warehouse operation requirements.
[0074] As a specific example, in step 5, it is determined whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and secondary sorting is performed on the finished products in the primary finished product warehouse according to the work order, as follows:
[0075] Step 5.1, the work order management system generates work orders based on customer orders or internal production plans. Each product is equipped with an RFID tag. The work order management system tracks the association between each product and its work order through an RFID reader and monitors the progress of work order execution in real time;
[0076] Step 5.2: If the quantity of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system controls the automatic stacker to take out the products required by the work order and place them on the roller line for further sorting;
[0077] Step 5.3: If the quantity of products in the primary finished product warehouse does not meet the work order requirements, the intelligent warehousing system will issue a work order instruction to the manufacturing department to replenish the current inventory to meet the needs of customer orders or internal production plans;
[0078] Step 5.4: After receiving the work order instruction, the manufacturing department manufactures and assembles the product according to the work order instruction. After the finished product is packed, it is marked with an RFID tag and registered in the work order management system;
[0079] Step 5.5: The finished products enter the primary finished product warehouse through the automated conveyor line, and the system automatically updates the inventory and associates it with the work order;
[0080] Step 5.6: The work order management system monitors inventory changes in real time and updates the work order execution progress according to the actual situation;
[0081] Step 5.7: When the number of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system starts the automatic stacker to take out the products for subsequent operations.
[0082] As a specific example, in step 6, the secondary sorted products are secondary packaged and stored in the finished product warehouse, as follows:
[0083] Step 6.1: The automatic packing system uses a robot to grab the primary finished product from the primary finished product warehouse and place it in the finished product box;
[0084] Step 6.2: The automatic packing system optimizes the placement strategy based on product size and box space to maximize packing efficiency;
[0085] Step 6.3: After packing is completed, the automatic packing system identifies each finished product box through the RFID tag and registers the finished product box into the work order management system. The finished product box is transported to the finished product inventory area through the automatic conveying system, and the inventory information is updated in real time.
[0086] As a specific example, in step 7, it is determined whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and the finished products in the finished product warehouse are sorted for shipment according to the work order, as follows:
[0087] Step 7.1, the intelligent warehousing system evaluates the latest delivery time of the product specified in the customer order and determines whether there is a sufficient quantity of customer products available for shipment by monitoring the finished product inventory;
[0088] Step 7.2: If the finished product inventory is insufficient to meet the delivery requirements of the customer's order, a production order is issued to the manufacturing department. At the same time, the system informs the manufacturing department of the latest production time of the product to ensure that production and assembly are completed before the delivery time required by the customer.
[0089] Step 7.3: When there are enough products in the finished goods inventory to meet the shipping conditions, the system starts the shipping sorting process. According to the customer order details, the corresponding products are taken out of the finished goods inventory and sorted in preparation for shipment.
[0090] As a specific example, in step 8, the sorted finished products are packed on pallets as follows:
[0091] Step 8.1: The intelligent warehousing system prepares pallets of appropriate sizes according to customer orders or shipping plans, and transports them to the packaging area through the conveyor system;
[0092] Step 8.2: The robot places the goods delivered from the finished product warehouse on the pallet, uses the A* algorithm to evaluate the shortest path and avoid obstacles, and uses the genetic algorithm to optimize the three-dimensional stacking of the products on the pallet;
[0093] Step 8.3: The robot simulates the stability assessment of the products on the pallet through the reinforcement learning algorithm and adjusts the stacking position of the products to reduce the instability factors during transportation.
[0094] Step 8.4: The intelligent warehousing system uses the computer vision algorithm YOLO to visually identify and detect each pallet to confirm the packaging integrity. Each pallet is attached with an RFID tag for subsequent inventory management and logistics tracking.
[0095] Step 8.5: After packaging is completed, the system moves the pallet to the shipping area or the waiting area of the logistics center, waiting for further warehousing and transportation arrangements to meet the customer's shipping requirements and schedule.
[0096] The present invention is further described below in conjunction with specific embodiments. The accompanying drawings are only used for exemplary descriptions and are only schematic diagrams rather than physical drawings, and should not be construed as limiting the present invention.
[0097] Example
[0098] Combination Figure 1 The present invention provides an electronic product intelligent storage management method, characterized in that it includes the following steps:
[0099] Step 1: The manufacturing department produces the product;
[0100] Step 2: The quality inspection system uses a machine vision system to inspect the product and feeds the inspection results back to the production process to improve production efficiency and product quality. The details are as follows:
[0101] Step 2.1: According to the order, the manufacturing department sends the new products to the warehouse automatic line;
[0102] Step 2.2: The visual inspection system obtains product images through the industrial camera installed at the front end of the warehouse automatic line;
[0103] Step 2.3, preprocessing the acquired product image;
[0104] Step 2.4: Use image processing and computer vision techniques to extract features from the preprocessed image;
[0105] Step 2.5: Based on the extracted features, use the deep learning model to perform defect detection and classification as follows:
[0106] Step 2.5.1, collect a large number of image datasets containing normal products and different types of defects;
[0107] Step 2.5.2, preprocess the images in the image dataset, including denoising, contrast enhancement and size standardization;
[0108] Step 2.5.3, build a CNN model, including convolutional layer, pooling layer, batch normalization layer and fully connected layer;
[0109] Step 2.5.4: Use the cross entropy loss function and Adam optimizer to train the model and adjust the hyperparameters to optimize the model performance.
[0110] Step 2.5.5, output the classification results of defect types;
[0111] Step 2.5.6: Evaluate the accuracy, precision, and recall of the model to ensure good performance on the validation and test sets.
[0112] Step 2.5.7: Deploy the trained model to the quality inspection system for defect detection and classification, and monitor and continuously optimize the model to adapt to new product changes and defect types.
[0113] Step 2.6: Based on the model output results, determine whether the product is qualified and what defects the unqualified products have;
[0114] Step 2.7: The quality inspection system records the quality inspection results of each product and performs data analysis as follows:
[0115] Step 2.7.1, collect the quality inspection results of the defect-detected products and store them in the central database; the product quality inspection results include structured data and unstructured data, where the structured data includes product ID, quality inspection time, defect type, and the unstructured data includes product images;
[0116] Step 2.7.2: Perform data cleaning to handle outliers, missing values, or erroneous values in the data, and extract useful features from the quality inspection data, including the frequency of each defect and the product quality inspection pass rate;
[0117] Step 2.7.3: Analyze the data using machine learning models, use the K-value clustering algorithm to group products according to quality inspection features, identify product groups with similar defect patterns, use the isolation forest algorithm to identify abnormal quality inspection results, and use cross-validation techniques to evaluate the performance of the model, and adjust the model hyperparameters to optimize the algorithm's predictive and generalization capabilities;
[0118] Step 2.7.4: Use real-time data stream processing technology to monitor and analyze the quality inspection process in real time to quickly discover and respond to potential problems;
[0119] Step 2.7.5: Use charts to display key indicators and trends of quality inspection data on the intelligent warehousing system interface to help decision makers understand the data and make corresponding management decisions.
[0120] Step 2.8: Feed the results of the quality inspection system back to the production process, and ensure the stability and continuous improvement of product quality on the production line through automated equipment adjustment and manual intervention.
[0121] Step 3: Primary packaging of qualified products, as follows:
[0122] Step 3.1: The automatic packing system receives qualified products that have passed quality inspection, and the robot uses the feature extraction and pattern recognition algorithm SURF to identify the position and direction of the product in the working area for subsequent grasping and placement;
[0123] Step 3.2: Use the inverse kinematics algorithm to calculate the motion parameters of each joint of the manipulator according to the target position and posture, use the A* algorithm to plan the motion path of the manipulator, and use the GJK algorithm to detect possible collisions during the movement of the manipulator in real time to ensure the safety of the operation;
[0124] Step 3.3: Use sensor data to monitor and adjust the gripping force and posture of the robot in real time to adapt to the characteristics of different products, and use model predictive control (MPC) to ensure the accuracy and stability of the gripping action.
[0125] Step 3.4: Use Kalman filtering to track targets in a dynamic environment, and control the robot to grab the product and move it to the target location based on visual feedback and sensor data;
[0126] Step 3.5: According to different product sizes and box spaces, use dynamic programming to optimize product placement strategies. Use a machine vision system combined with the object detection model YOLO based on the convolutional neural network CNN to detect and identify products in real time to ensure the correct placement of products and avoid position errors or overlaps.
[0127] Step 3.6: Use the real-time data provided by sensors and vision systems to detect and provide feedback on abnormalities during the packing process;
[0128] Step 3.7: After placing the products, seal the boxes.
[0129] Step 4: Store the qualified products in primary packaging in the primary finished product warehouse, as follows:
[0130] Step 4.1: Collect historical order data of the product, including model, size, weight, and sales frequency information;
[0131] Step 4.2: Extract product features, including size, weight, classification, sales trend, warehouse temperature and humidity, as input features of the model;
[0132] Step 4.3: Use historical order data to train a deep neural network (DNN) model so that it can accurately predict the optimal storage location and cargo allocation for different products.
[0133] Step 4.4: Deploy the trained deep neural network DNN model to the real-time system to monitor new orders and inventory changes in real time, predict and optimize product location allocation and storage strategies based on real-time data, ensure timely fulfillment of order requirements and optimize warehouse space utilization;
[0134] Step 4.5: Dynamically adjust the parameters and strategies of the machine learning model based on the data feedback from actual operation to optimize the storage efficiency and operation process of the product;
[0135] Step 4.6: Use evaluation indicators such as picking time, inventory turnover rate, space utilization, etc. to evaluate the effectiveness of the machine learning model. Adjust and optimize the machine learning model based on the evaluation results to continuously improve the product's storage management strategy to meet changing market demands and warehouse operation requirements.
[0136] Step 5: Determine whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and perform secondary sorting on the finished products in the primary finished product warehouse according to the work order, as follows:
[0137] Step 5.1, the work order management system generates work orders based on customer orders or internal production plans. Each product is equipped with an RFID tag. The work order management system tracks the association between each product and its work order through an RFID reader and monitors the progress of work order execution in real time;
[0138] Step 5.2: If the quantity of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system controls the automatic stacker to take out the products required by the work order and place them on the roller line for further sorting;
[0139] Step 5.3: If the quantity of products in the primary finished product warehouse does not meet the work order requirements, the intelligent warehousing system will issue a work order instruction to the manufacturing department to replenish the current inventory to meet the needs of customer orders or internal production plans;
[0140] Step 5.4: After receiving the work order instruction, the manufacturing department manufactures and assembles the product according to the work order instruction. After the finished product is packed, it is marked with an RFID tag and registered in the work order management system;
[0141] Step 5.5: The finished products enter the primary finished product warehouse through the automated conveyor line, and the system automatically updates the inventory and associates it with the work order;
[0142] Step 5.6: The work order management system monitors inventory changes in real time and updates the work order execution progress according to the actual situation;
[0143] Step 5.7: When the number of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system starts the automatic stacker to take out the products for subsequent operations.
[0144] Step 6: Secondary pack the secondary sorted products and store them in the finished product warehouse, as follows:
[0145] Step 6.1: The automatic packing system uses a robot to grab the primary finished product from the primary finished product warehouse and place it in the finished product box;
[0146] Step 6.2: The automatic packing system optimizes the placement strategy based on product size and box space to maximize packing efficiency;
[0147] Step 6.3: After packing is completed, the automatic packing system identifies each finished product box through the RFID tag and registers it in the work order management system. The finished product box is transported to the finished product inventory area through the automatic conveying system, and the inventory information is updated in real time.
[0148] Step 7: Determine whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and sort the finished products in the finished product warehouse for shipment according to the work order, as follows:
[0149] Step 7.1, the intelligent warehousing system evaluates the latest delivery time of the product specified in the customer order and determines whether there is a sufficient quantity of customer products available for shipment by monitoring the finished product inventory;
[0150] Step 7.2: If the finished product inventory is insufficient to meet the delivery requirements of the customer's order, a production order is issued to the manufacturing department. At the same time, the system informs the manufacturing department of the latest production time of the product to ensure that production and assembly are completed before the delivery time required by the customer.
[0151] Step 7.3: When there are enough products in the finished goods inventory to meet the shipping conditions, the system starts the shipping sorting process. According to the customer order details, the corresponding products are taken out of the finished goods inventory and sorted in preparation for shipment.
[0152] Step 8: Pack the sorted finished products on pallets as follows:
[0153] Step 8.1: The intelligent warehousing system prepares pallets of appropriate sizes according to customer orders or shipping plans, and transports them to the packaging area through the conveyor system;
[0154] Step 8.2: The robot places the goods delivered from the finished product warehouse on the pallet, uses the A* algorithm to evaluate the shortest path and avoid obstacles, and uses the genetic algorithm to optimize the three-dimensional stacking of the products on the pallet to maximize space utilization while ensuring stability.
[0155] Step 8.3: The robot simulates the stability assessment of the products on the pallet through the reinforcement learning algorithm and adjusts the stacking position of the products to reduce the instability factors during transportation.
[0156] Step 8.4: The intelligent warehousing system uses the computer vision algorithm YOLO to visually identify and detect each pallet to confirm the packaging integrity. Each pallet is attached with an RFID tag for subsequent inventory management and logistics tracking.
[0157] Step 8.5: After packaging is completed, the system moves the pallet to the shipping area or the waiting area of the logistics center, waiting for further warehousing and transportation arrangements to meet the customer's shipping requirements and schedule.
[0158] Step 9: Ship the packaged finished products out of the warehouse.
[0159] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. An intelligent storage management method for electronic products, characterized in that: The following steps are involved: Step 1: The manufacturing department produces the product; Step 2: The quality inspection system uses a machine vision system to inspect the product and feeds the inspection results back to the production process to improve production efficiency and product quality. Step 3: Primary packaging of qualified products; Step 4: Store the qualified products in primary packaging in the primary finished product warehouse; Step 5: Determine whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and perform secondary sorting on the finished products in the primary finished product warehouse according to the work order; Step 6: Secondarily package the secondary sorted products and store them in the finished product warehouse; Step 7: Determine whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and sort the finished products in the finished product warehouse for shipment according to the work order; Step 8: Pack the sorted finished products on pallets; Step 9: Ship the packaged finished products out of the warehouse.
2. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 2, the quality inspection system uses a machine vision system to inspect the product and feeds the inspection results back to the production process to improve production efficiency and product quality, as follows: Step 2.1: According to the order, the manufacturing department sends the new products to the warehouse automatic line; Step 2.2: The visual inspection system obtains product images through the industrial camera installed at the front end of the warehouse automatic line; Step 2.3, preprocessing the acquired product image; Step 2.4: Use image processing and computer vision techniques to extract features from the preprocessed image; Step 2.5: Based on the extracted features, use the deep learning model to perform defect detection and classification; Step 2.6: Based on the model output results, determine whether the product is qualified and what defects the unqualified products have; Step 2.7: The quality inspection system records the quality inspection results of each product and performs data analysis; Step 2.8: Feed the results of the quality inspection system back to the production process, and improve the product quality on the production line through automated equipment adjustments and manual intervention.
3. The electronic product intelligent storage management method according to claim 2, characterized in that: In step 2.5, based on the extracted features, a deep learning model is used for defect detection and classification, as follows: Step 2.5.1, collect image datasets containing normal products and different types of defects; Step 2.5.2, preprocess the images in the image dataset, including denoising, contrast enhancement and size standardization; Step 2.5.3, build a CNN model, including convolutional layer, pooling layer, batch normalization layer and fully connected layer; Step 2.5.4: Use the cross entropy loss function and Adam optimizer to train the model and adjust the hyperparameters to optimize the model performance. Step 2.5.5, output the classification results of defect types; Step 2.5.6: Evaluate the accuracy, precision, and recall metrics of the model; Step 2.5.7: Deploy the trained model to the quality inspection system for defect detection and classification, and monitor and continuously optimize the model to adapt to new product changes and defect types.
4. The electronic product intelligent storage management method according to claim 2, characterized in that: In step 2.7, the quality inspection system records the quality inspection results of each product and performs data analysis as follows: Step 2.7.1, collect the quality inspection results of the defect-detected products, including structured data and unstructured data, and store them in the central database; Step 2.7.2: Perform data cleaning to handle outliers, missing values, or erroneous values in the data, and extract useful features from the quality inspection data, including the frequency of each defect and the product quality inspection pass rate; Step 2.7.3: Analyze the data using machine learning models, use the K-value clustering algorithm to group products according to quality inspection features, identify product groups with similar defect patterns, use the isolation forest algorithm to identify abnormal quality inspection results, and use cross-validation techniques to evaluate the performance of the model, and adjust the model hyperparameters to optimize the algorithm's predictive and generalization capabilities; Step 2.7.4: Use real-time data stream processing technology to monitor and analyze the quality inspection process in real time to identify and respond to potential problems; Step 2.7.5: Use charts to display key indicators and trends of quality inspection data on the intelligent warehousing system interface to help decision makers understand the data and make corresponding management decisions.
5. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 3, the qualified products are packaged in primary packaging as follows: Step 3.1: The automatic packing system receives qualified products that have passed quality inspection. The robot uses the feature extraction and pattern recognition algorithm SURF to identify the position and direction of the product in the working area for subsequent grasping and placement. Step 3.2: Use the inverse kinematics algorithm to calculate the motion parameters of each joint of the manipulator according to the target position and posture, use the A* algorithm to plan the motion path of the manipulator, and use the GJK algorithm to detect possible collisions during the movement of the manipulator in real time; Step 3.3, use sensor data to monitor and adjust the gripping force and posture of the robot in real time to adapt to the characteristics of different products, and use model predictive control MPC; Step 3.4: Use Kalman filtering to track targets in a dynamic environment, and control the robot to grab the product and move it to the target location based on visual feedback and sensor data; Step 3.5: According to different product sizes and box spaces, use dynamic programming to optimize product placement strategies. Use a machine vision system combined with the object detection model YOLO based on the convolutional neural network CNN to detect and identify products in real time to ensure correct product placement. Step 3.6: Use the real-time data provided by sensors and vision systems to detect and provide feedback on abnormalities during the packing process; Step 3.7: After placing the products, seal the boxes.
6. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 4, the qualified products in primary packaging are stored in the primary finished product warehouse, as follows: Step 4.1: Collect historical order data of the product, including model, size, weight, and sales frequency information; Step 4.2: Extract product features, including size, weight, classification, sales trend, warehouse temperature and humidity, as input features of the model; Step 4.3: Use historical order data to train a deep neural network (DNN) model to predict the optimal storage location and cargo allocation for different products. Step 4.4: Deploy the trained deep neural network DNN model to the real-time system to monitor new orders and inventory changes in real time, predict and optimize product location allocation and storage strategies based on real-time data, ensure timely fulfillment of order requirements and optimize warehouse space utilization; Step 4.5: Dynamically adjust the parameters and strategies of the machine learning model based on the data feedback from actual operation to optimize the storage efficiency and operation process of the product; Step 4.6: Use evaluation indicators to evaluate the effectiveness of the machine learning model, and adjust and optimize the machine learning model based on the evaluation results to improve the product's storage location management strategy to meet changing market demands and warehouse operation requirements.
7. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 5, it is determined whether the quantity of products in the primary finished product warehouse meets the secondary packaging conditions, and secondary sorting is performed on the finished products in the primary finished product warehouse according to the work order, as follows: Step 5.1, the work order management system generates work orders based on customer orders or internal production plans. Each product is equipped with an RFID tag. The work order management system tracks the association between each product and its work order through an RFID reader and monitors the progress of work order execution in real time; Step 5.2: If the quantity of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system controls the automatic stacker to take out the products required by the work order and place them on the roller line for further sorting; Step 5.3: If the quantity of products in the primary finished product warehouse does not meet the work order requirements, the intelligent warehousing system will issue a work order instruction to the manufacturing department to replenish the current inventory to meet the needs of customer orders or internal production plans; Step 5.4: After receiving the work order instruction, the manufacturing department manufactures and assembles the product according to the work order instruction. After the finished product is packed, it is marked with an RFID tag and registered in the work order management system; Step 5.5: The finished products enter the primary finished product warehouse through the automated conveyor line, and the system automatically updates the inventory and associates it with the work order; Step 5.6: The work order management system monitors inventory changes in real time and updates the work order execution progress according to the actual situation; Step 5.7: When the number of products in the primary finished product warehouse meets the work order requirements, the intelligent warehousing system starts the automatic stacker to take out the products for subsequent operations.
8. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 6, the secondary sorted products are secondary packaged and stored in the finished product warehouse, as follows: Step 6.1: The automatic packing system uses a robot to grab the primary finished product from the primary finished product warehouse and place it in the finished product box; Step 6.2: The automatic packing system optimizes the placement strategy based on product size and box space to maximize packing efficiency; Step 6.3: After packing is completed, the automatic packing system identifies each finished product box through the RFID tag and registers the finished product box into the work order management system. The finished product box is transported to the finished product inventory area through the automatic conveying system, and the inventory information is updated in real time.
9. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 7, it is determined whether the quantity of products in the finished product warehouse meets the shipping conditions of the work order, and the finished products in the finished product warehouse are sorted for shipment according to the work order, as follows: Step 7.1, the intelligent warehousing system evaluates the latest delivery time of the product specified in the customer order and determines whether there is a sufficient quantity of customer products available for shipment by monitoring the finished product inventory; Step 7.2: If the finished product inventory is insufficient to meet the delivery requirements of the customer's order, a production order is issued to the manufacturing department. At the same time, the system informs the manufacturing department of the latest production time of the product to ensure that production and assembly are completed before the delivery time required by the customer. Step 7.3: When there are enough products in the finished goods inventory to meet the shipping conditions, the system starts the shipping sorting process. According to the customer order details, the corresponding products are taken out of the finished goods inventory and sorted in preparation for shipment.
10. The electronic product intelligent storage management method according to claim 1, characterized in that: In step 8, the sorted finished products are palletized as follows: Step 8.1: The intelligent warehousing system prepares pallets of appropriate sizes according to customer orders or shipping plans, and transports them to the packaging area through the conveyor system; Step 8.2: The robot places the goods delivered from the finished product warehouse on the pallet, uses the A* algorithm to evaluate the shortest path and avoid obstacles, and uses the genetic algorithm to optimize the three-dimensional stacking of the products on the pallet; Step 8.3: The robot simulates the stability assessment of the products on the pallet through the reinforcement learning algorithm and adjusts the stacking position of the products to reduce the instability factors during transportation. Step 8.4: The intelligent warehousing system uses the computer vision algorithm YOLO to visually identify and detect each pallet to confirm the packaging integrity. Each pallet is attached with an RFID tag for subsequent inventory management and logistics tracking. Step 8.5: After packaging is completed, the system moves the pallet to the shipping area or the waiting area of the logistics center, waiting for further warehousing and transportation arrangements to meet the customer's shipping requirements and schedule.
Citation Information
Cited By
Product packaging management system and method based on intelligent specification matching
CN121094743A
