A goods dispatching and distribution management system suitable for a stacker vertical warehouse

Abstract

This invention discloses a cargo scheduling and distribution management system suitable for stacker crane automated storage and retrieval systems (AS / RS), relating to the field of logistics and warehousing technology. It includes an intelligent scheduling module: employing a multi-algorithm fusion and dynamic switching mechanism to plan stacker crane paths and allocate tasks based on orders and warehouse conditions; a precise inventory management module: utilizing multi-sensor fusion and intelligent recognition technology, combined with environmental adaptation and intelligent auditing to ensure accurate inventory data; a system integration module: achieving multi-system integration through standardized interfaces and middleware, possessing cluster deployment, fault handling, and interface management mechanisms; an equipment fault response module; and an energy management module: optimizing energy utilization based on an energy-efficiency balance model combined with hybrid energy and intelligent switching technology. Through the close collaboration of multiple modules and the comprehensive application of innovative technologies, this invention achieves efficient, intelligent, and reliable operation of the stacker crane AS / RS cargo scheduling and distribution management system.

Description

Technical Field

[0001] This invention relates to the field of logistics and warehousing technology, specifically to a cargo dispatching and distribution management system suitable for stacker crane automated warehouses. Background Technology

[0002] In modern logistics warehousing, stacker cranes play a crucial role. However, traditional management methods rely on manual experience, which has many shortcomings. In path planning, manual methods struggle to determine the optimal route, leading to time-consuming and energy-intensive stacker crane operations. Inventory management lacks precision and real-time capabilities, easily resulting in stockpiling or shortages, impacting business operations. System integration is low, with poor data exchange with external systems creating information silos. Equipment failure handling is inadequate, lacking intelligent detection and early warning systems, easily causing operational chaos during malfunctions. Energy management is inefficient and lacks flexible control, resulting in energy waste. Personnel operation and training systems are inadequate, with unfriendly interfaces and outdated training updates. Furthermore, the lack of effective artificial intelligence prediction leads to low collaborative operation efficiency, unstable communication, and poor task allocation and traffic management.

[0003] In view of the above, this application is hereby submitted. Summary of the Invention

[0004] The purpose of this invention is to provide a cargo scheduling and distribution management system suitable for stacker crane automated warehouses, so as to solve the problem of low cargo scheduling and distribution efficiency in existing stacker crane automated warehouses mentioned in the background art.

[0005] To address the aforementioned technical issues, this invention provides a goods dispatching and distribution management system suitable for stacker crane automated warehouses (AS / RS), comprising: an intelligent dispatching module: employing a multi-algorithm fusion and dynamic switching mechanism to plan stacker crane paths and assign tasks based on orders and warehouse conditions, and featuring special goods handling strategies; a precise inventory management module: utilizing multi-sensor fusion and intelligent recognition technology, combined with environmental adaptation and intelligent auditing to ensure accurate inventory data; a system integration module: achieving multi-system integration through standardized interfaces and middleware, possessing cluster deployment, fault handling, and interface management mechanisms; an equipment fault response module: achieving fault detection and early warning through multi-sensor and machine learning algorithms, and optimizing emergency dispatching through distributed computing and real-time monitoring; an energy management module: optimizing energy utilization based on an energy-efficiency balance model combined with hybrid energy and intelligent switching technology; a personnel operation and training assistance module: providing a personalized safety customization interface and a high-quality training system; an artificial intelligence prediction module: integrating multi-source data and using visual multi-dimensional interactive tools to assist decision-making; and an automated warehouse and robot collaborative operation module: employing redundant communication networks and dynamic task management strategies to ensure efficient collaboration.

[0006] Furthermore, the intelligent scheduling module integrates multiple algorithms, including path optimization algorithms such as A* or Dijkstra's algorithm, with task allocation algorithms. The dynamic switching mechanism switches between conventional and fast scheduling algorithms based on order volume and computing resource status. Meanwhile, the special cargo handling strategy includes a special cargo database, its real-time updates, and the formulation of targeted scheduling rules.

[0007] Furthermore, the multi-sensor fusion of the precision inventory management module uses Kalman filtering or fuzzy logic algorithms to process sensor data such as RFID and laser scanning. The environment adaptively adjusts the data fusion weights according to regional differences, and the intelligent review includes a combination of artificial intelligence pre-screening and manual review.

[0008] Furthermore, the middleware of the system integration module adopts an enterprise service bus, the cluster deployment has node fault isolation and self-healing functions, and the interface management includes a version compatibility mechanism and a multi-system compatibility dynamic monitoring platform.

[0009] Furthermore, the machine learning algorithm of the equipment fault response module adopts a hybrid intelligent scheduling algorithm that integrates genetic algorithm and A* algorithm. The distributed computing architecture accelerates the data processing of the incremental learning algorithm, and the real-time monitoring adopts an event-driven architecture and intelligent caching technology to optimize the generation of scheduling schemes.

[0010] Furthermore, the energy management module's energy-efficiency balance model, combined with a real-time order correction mechanism, adjusts the stacker crane's energy strategy, and the hybrid energy intelligent switching system includes voltage stabilization filtering and pre-switching buffer functions.

[0011] Furthermore, the personalized security customization interface of the personnel operation and training assistance module is equipped with a security audit mechanism, and the automatic update mechanism of the training system includes a manual audit step to ensure training quality.

[0012] Furthermore, the multi-source data fusion of the artificial intelligence prediction module includes an external data quality assessment and cleaning system, and the visualization multi-dimensional interactive tool uses technologies such as pivot tables and interactive dashboards to achieve multi-dimensional analysis and display.

[0013] Furthermore, the redundant communication network between the automated warehouse and the robot collaborative operation module adopts double buffering technology and a seamless switching mechanism to ensure data transmission. The dynamic task management strategy adjusts the robot task priority and path planning in real time according to the urgency of the task and the equipment status.

[0014] Compared with the prior art, the beneficial effects of the present invention are:

[0015] 1. The intelligent scheduling module uses the A* algorithm to plan routes and allocate tasks. Taking into account multiple factors, it accurately determines the stacker crane's path, reducing travel distance and time. Faced with a large volume of orders, it can rationally schedule tasks based on urgency, enabling rapid order processing during peak e-commerce promotions. The automated warehouse and robot collaborative operation module optimizes the communication and management system, using algorithms to reduce robot conflicts and waiting times, ensuring smooth operations even during busy periods, accelerating goods handling, and improving operational efficiency.

[0016] 2. The system integration module leverages the ESB middleware to achieve efficient interaction and integration with external systems, clearly defining interfaces and specifications to enable information sharing and collaboration, such as synchronizing inventory data with the ERP system and optimizing resource allocation. Its robust interface management and testing mechanisms ensure compatibility during upgrades, reduce maintenance cost risks, and enhance system performance.

[0017] 3. The energy management module uses EMS to monitor and analyze the stacker crane's energy consumption, adjusts parameters according to operation and equipment status, combines renewable energy and builds a balance model, and adopts flexible strategies based on order urgency and energy reserves, such as reducing speed and using solar energy during off-peak hours, and appropriately relaxing energy consumption restrictions during emergencies, to achieve efficient energy use and rational allocation; and optimize resource utilization.

[0018] 4. The AI ​​prediction module uses multiple algorithms to predict the demand for goods entering and leaving the warehouse. The AI ​​prediction module integrates multi-source data and uses visualization and multi-dimensional interactive tools to assist decision-making. In the process of integrating multi-source data, it involves mining and analyzing industrial data, such as order data and inventory data. By mining the relationships and patterns between the data, and then using visualization tools to display them to assist decision-making, this is a typical application of industrial data mining and analysis; making predictions accurate and helping to make scientific decisions. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a cargo dispatching and distribution management system suitable for stacker crane automated warehouses. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1:

[0022] Suppose a large food and beverage manufacturing company has a wide variety of products and significant needs for the storage and distribution of raw materials and finished products. The company owns a modern stacker crane automated storage and retrieval system (AS / RS) warehouse, covering an area of ​​8,000 square meters with racking heights of 12 meters and a total of 8,000 storage locations. The company processes an average of 150 orders per day, each order containing 5-100 different items, covering various food raw materials, packaging materials, and finished beverages and food products.

[0023] Please see Figure 1 This invention provides a technical solution: a goods dispatching and distribution management system suitable for stacker crane automated warehouses, including an intelligent dispatching module. The real-time path optimization algorithm of the intelligent dispatching module adopts either the A* algorithm or the Dijkstra algorithm, and reduces algorithm complexity through pruning operations and heuristic function optimization. Simultaneously, it employs high-performance servers or distributed computing architectures to enhance computing resources to handle large-scale data processing. Furthermore, the intelligent dispatching module also includes a dynamic algorithm switching mechanism. When the order volume exceeds a certain threshold and computational delays begin to occur, it automatically switches to a simplified fast dispatching algorithm, initially allocating urgent orders to ensure goods can be moved as quickly as possible. Subsequently, it gradually optimizes paths using idle computing resources and establishes task connections. The transition algorithm comprehensively considers the stacker crane location in the initial task allocation, the completed handling progress, and the optimized path planning during algorithm switching to ensure a smooth task transition and reduce delays. It also includes a special cargo database, which identifies special cargo types before task allocation and then formulates special scheduling strategies for them based on their special requirements. For fragile items, bumpy sections are avoided during path planning, the acceleration and deceleration rates of the stacker crane are limited, and priority is given to assigning them to equipment corresponding to experienced stacker crane operators. A real-time monitoring and updating mechanism for the special cargo database is established, and information sharing channels are established with suppliers, R&D departments, etc. Once there is information on new special cargo or changes in special requirements, the database is updated immediately, and the scheduling strategy is adjusted and optimized simultaneously.

[0024] 1. Path planning and task allocation:

[0025] Stacker crane A is located at coordinates XA=5, YA=3, ZA=4; stacker crane B is located at XB=10, YB=6, ZB=5; the target storage location coordinates are XT=15, YT=8, ZT=7. Moving one unit horizontally costs 1.5 units, and moving one unit vertically costs 3 units.

[0026] According to the A* algorithm formula, A= First, calculate the distance from stacker crane A to the target storage location:

[0027] ;

[0028] Assume we first move from level A to... ,but Increase by 1.5, then move vertically to... ,but Increase ,at this time , (This is only a preliminary calculation; actual calculations need to consider every step of the entire path.) Through this calculation, the system can compare different paths. The value determines the optimal path.

[0029] Assume there are three stacker cranes. The straight-line distance between stacker crane A and the target goods is: rice,

[0030] The distance of stacker crane C is: rice,

[0031] The distance of stacker crane C is: rice,

[0032] Based on the principle of proximity, it should be allocated to stacker crane B first.

[0033] Assuming order urgency is categorized into high (0.8), medium (0.5), and low (0.2), the current order urgency is high. There are 5 stacker cranes, and their load conditions are as follows: Stacker crane 1 has assigned tasks. Stacker crane 2 workload Stacker crane 3 workload Stacker crane 4 workload Stacker crane 5 workload Average workload Calculate the deviation , , , , Preferred The corresponding stacker crane is 3, but since stacker crane B is closer and more urgent, it is still prioritized to be assigned to stacker crane B after comprehensive consideration.

[0034] 2. Special cargo handling:

[0035] A special cargo scheduling priority model can be established, assuming the fragility of the special cargo is 1. (Values ​​range from 0 to 1, with values ​​closer to 1 indicating greater fragility), the degree of perishability is... (Values ​​range from 0 to 1, with values ​​closer to 1 indicating greater perishability) Fragility of a specific type of cargo Degree of perishability Set priority for special cargo scheduling (The weighting coefficients are set based on the company's experience), then ,according to Special cargo tasks are prioritized and processed first, assigned to equipment by experienced operators. Bumpy road sections are avoided during route planning. Assuming bumpy road sections comprise 8% of the warehouse floor area, primarily located along the warehouse edges near the loading / unloading area, the acceleration and deceleration rates of the stacker crane are limited, with acceleration not exceeding 0.3 m / s. 2 The deceleration rate shall not exceed 0.2 m / s. 2 .

[0036] A goods dispatching and distribution management system suitable for stacker crane automated warehouses includes a precision inventory management module. This module incorporates RFID and laser scanning sensors and employs a data fusion algorithm based on Kalman filtering or fuzzy logic to fuse data from different sensors. A sensor calibration mechanism is established to periodically calibrate the sensors to reduce data errors. Furthermore, an environmental adaptation module is added to the data fusion algorithm. Through additional environmental sensors, such as dust concentration sensors and metal detectors, it monitors environmental changes in real time. When environmental interference exceeds a certain range, it automatically adjusts the weights or parameters of the data fusion algorithm. The threshold setting and judgment logic of the environmental adaptation module are optimized, and the system is integrated with the warehouse... The layout and operation area division include setting differentiated tolerance levels for environmental changes in different areas. It also involves strengthening the training of the intelligent identification system, collecting more subtle feature data on similar-looking goods, and adopting a more refined deep learning model structure, such as introducing an attention mechanism, to enable the model to focus on key differentiating features of the goods. Simultaneously, a manual review mechanism is established; when the similarity of the identification results is high and there are doubts, the manual review process is automatically triggered to ensure identification accuracy. Furthermore, artificial intelligence is introduced to assist manual review, using machine learning models to pre-screen and classify goods suspected of being misidentified, prioritizing the review of high-risk goods, such as high-value, low-quantity goods, to improve review efficiency.

[0037] 1. Data Acquisition and Fusion:

[0038] A batch of beverage boxes has an external dimension of length. Meters, width meters, height Meters. Laser scanning sensors measure the weight of goods. kilograms, the theoretical weight recorded in the database associated with the RFID sensor is [weight missing]. kilogram.

[0039] Let the state equation of the measurement system be: (in , , , Zero-mean Gaussian noise, covariance matrix The measurement equation is: ( For the measurement matrix, It is zero-mean Gaussian noise, and the covariance matrix is... First, initialize the state estimation. Covariance Matrix According to the Kalman filter formula:

[0040] ;

[0041] ,

[0042] To obtain a more accurate weight estimate kilogram.

[0043] 2. Environmental adaptation and identification optimization:

[0044] When dust concentration (Exceeding the threshold) ), metal interference intensity microtesla (exceeding the threshold) (micro-Tesla), let environmental impact factors be defined. (The coefficients are determined based on experimental data), then .according to The sensor weights are adjusted accordingly. Under normal circumstances, the weights of the laser scanning sensor data are: The weight of RFID sensor data is The adjusted weights of the laser scanning sensor data are: The weight of RFID sensor data is .

[0045] Assuming a Convolutional Neural Network (CNN) is used for cargo recognition, the network structure includes three convolutional layers, two pooling layers, and two fully connected layers. The input cargo image is processed by convolutional layers to extract features, pooling layers for dimensionality reduction, and fully connected layers for classification. Cosine similarity formula is used to calculate the recognition similarity. Suppose that the feature vectors of a certain cargo and the target cargo are... , ,but

[0046] , ,

[0047] , ,

[0048] If the similarity exceeds This will automatically trigger a manual review process.

[0049] A goods dispatching and distribution management system suitable for stacker crane automated warehouses includes a system integration module. The middleware of the system integration module adopts Enterprise Service Bus (ESB) middleware, and its modular design follows the principle of low coupling, strictly defining interface standards and data interaction specifications between modules, and establishing communication mechanisms and coordination frameworks between modules. Furthermore, the middleware is deployed in a cluster, employing load balancing and caching technologies. When the network is unstable, data is temporarily stored in the cache, and automatically resent after the network recovers. Simultaneously, the middleware performance is periodically stress-tested, and the cluster configuration is dynamically adjusted based on the test results. A node fault isolation and self-healing mechanism is established for the middleware cluster; when a node fails, it is automatically isolated. Tasks are dynamically assigned to other healthy nodes, and a self-healing program is initiated to diagnose and repair faulty nodes. This also includes establishing an interface version management mechanism to ensure that new and old interfaces can coexist and remain compatible for a period of time during module function upgrades. When interface standards are modified, the relevant module development teams are notified in advance, and detailed interface change documentation and adaptation sample code are provided. Simultaneously, automated interface compatibility testing tools are developed to conduct comprehensive compatibility testing after module upgrades, promptly identifying and resolving issues. Furthermore, a multi-system interface version compatibility dynamic monitoring platform is established to monitor changes in interface versions and data interaction across systems in real time, providing early warnings of potential compatibility issues and offering automated compatibility test scripts and repair suggestions.

[0050] 1. Data Interaction and Modular Design:

[0051] The inventory management module interacts with the ERP system to exchange data such as inventory quantity and inventory cost. Assuming that within a certain time period, the inventory management module transmits data to the ERP system... Successful transmission Next, the reliability of data interaction By monitoring This value ensures the stability of data interaction. The value is lower than the set threshold (e.g.) If so, check for problems such as data transmission lines and interfaces.

[0052] 2. Middleware Deployment and Management:

[0053] Each middleware node can initially process The system simulates requests per second, using Nginx as a load balancer to evenly distribute data requests across four middleware nodes. In the stress test, it simulates... Requests per second; if the average response time exceeds 400 milliseconds, add a node. Let the node... The load is requests / second, load balancing For example, the actual loads of the four nodes are respectively , , , ,but .like An excessively large value indicates an unbalanced load, requiring adjustment of the load balancing strategy.

[0054] Establish an interface version management mechanism, assuming that the inventory management module upgrade allows the new and old interfaces to coexist and remain compatible for four months. When modifying interface standards, notify the relevant module development teams three weeks in advance, providing detailed interface change documentation and adaptation sample code. Develop automated interface compatibility testing tools to conduct comprehensive compatibility testing after module upgrades. Let the interface compatibility index be [insert index here]. Based on a comprehensive evaluation of factors such as interface parameter matching degree and data format consistency, if The value is lower than the set threshold (e.g.) If the result is negative, it indicates a compatibility issue that requires further inspection and modification.

[0055] A goods dispatching and distribution management system suitable for stacker crane automated warehouses includes an equipment failure response module. The machine learning algorithm of this module employs a hybrid intelligent scheduling algorithm combining genetic algorithms and A* algorithms to improve the adaptability and stability of emergency dispatch strategies. An adaptive adjustment mechanism is established to automatically adjust algorithm parameters or switch algorithm strategies based on changes in the warehouse environment. Furthermore, an incremental learning algorithm is used to update the failure prediction model. When new failure cases occur, their data are promptly added to the training set, enabling the model to continuously learn new failure characteristics. Simultaneously, a failure case database is established to record and analyze newly occurring failures in detail, allowing for timely model adjustments and the development of corresponding countermeasures. A distributed computing architecture is also employed. To accelerate data processing for incremental learning algorithms, new fault data is distributed across multiple computing nodes for parallel processing, improving model update speed. This also includes adding real-time monitoring and dynamic adjustment mechanisms to the fault emergency scheduling strategy. During emergency scheduling, real-time changes in equipment and the environment within the warehouse are continuously monitored. When unexpected situations such as backup equipment failure or channel obstruction are detected, the scheduling plan is immediately re-planned, utilizing other available resources or adjusting transport routes to ensure the continued smooth dispatch of goods. Furthermore, the real-time monitoring information processing flow is optimized by adopting an event-driven architecture and intelligent caching technology. When key information within the warehouse, such as channel status and location, changes, the corresponding scheduling plan adjustment procedure is immediately triggered, and new scheduling paths are quickly generated using cached data.

[0056] 1. Data Acquisition and Algorithm Application:

[0057] The motor temperature data collected by the temperature sensor is , , The vibration amplitude data collected by the vibration sensor is... millimeters The current data collected by the current sensor is in millimeters, etc. , wait.

[0058] Let the fitness function of the genetic algorithm be... Assuming During training, if accuracy False alarm rate ,but:

[0059] By continuously iterating and updating the parameter vector Make To achieve optimality, thereby optimizing the parameters of the fault detection model.

[0060] 2. Model Updates and Emergency Handling:

[0061] Incremental learning algorithms are used to update the fault prediction model, assuming the accuracy of the fault prediction model evaluation index is... In the initial model, , , , ,but When new failure cases emerge and their data is added to the training set, the accuracy is recalculated. If the accuracy improves to [a certain level], [the accuracy is then considered]. This indicates that the model performance has been improved.

[0062] During emergency dispatch, assuming normal dispatch time Minutes, emergency dispatch completion time Minutes, then emergency dispatch efficiency By monitoring The effectiveness of emergency dispatch strategies is evaluated. If the value is low, the emergency dispatch plan should be adjusted, such as optimizing resource allocation and improving route planning.

[0063] A goods dispatching and distribution management system suitable for stacker crane automated warehouses includes an energy management module. This module further includes establishing an energy-efficiency balance model to dynamically adjust the stacker crane's energy optimization strategy based on order urgency and energy reserves. When orders are urgent, energy consumption restrictions are appropriately relaxed to ensure rapid goods entry and exit from the warehouse. When order volume is low, energy optimization efforts are increased, reducing operating speed and power. Simultaneously, by predicting order volume and energy demand, the stacker crane's operation plan is scheduled in advance, making energy utilization more rational. Furthermore, a hybrid energy storage and intelligent switching system is adopted to combine renewable energy sources such as solar and wind power with traditional electricity. It is equipped with energy storage devices, such as battery packs. When renewable energy is abundant, it prioritizes the use of renewable energy and stores excess electricity. When renewable energy is insufficient, it automatically switches to traditional power supply and dynamically adjusts the power usage strategy based on energy storage conditions and energy prices. It also introduces a real-time order correction mechanism to supplement the energy-efficiency balance model. Combining the actual operation of the warehouse and the order processing progress, it corrects the predicted value of order urgency in real time to ensure accurate and reasonable energy allocation. It also includes optimizing the control logic and hardware performance of the hybrid energy switching device, adding voltage stabilization and filtering functions, and setting up pre-switching detection and buffering mechanisms to ensure a smooth transition during the energy switching process.

[0064] 1. Monitoring and parameter adjustment:

[0065] The power consumption data of the stacker crane within a certain period of time is as follows: kilowatt-hours kilowatt-hours, etc., operating speed is meters per second, load weight is kilogram.

[0066] When order volume , , At meters per second, according to the formula , meters per second, that is, the adjusted speed is approximately meters per second.

[0067] Assuming a model relating energy consumption and workload The results were obtained by fitting experimental data. , , Then when , , , Hour (minute) The energy consumption per kilowatt-hour can be estimated under different operating conditions based on this model, and operating parameters can be adjusted accordingly.

[0068] Assuming the area of ​​the solar panels on the top of the warehouse square meters, photoelectric conversion efficiency Local solar intensity kilowatts per square meter, time If the solar power generation is [amount] per hour, then the solar power generation is [amount].

[0069] Kilowatt-hours can be used preferentially and excess electricity can be stored when solar energy is plentiful.

[0070] 2. Energy Strategy and Switching Optimization:

[0071] When the urgency of the order , kilowatt-hours per hour At that time, according to the formula:

[0072] , Energy consumption restrictions should be appropriately relaxed to ensure that goods can enter and leave the warehouse quickly.

[0073] Suppose we predict the order volume for the next week using the ARIMA model, and let the order volume sequence be... Establish an ARIMA(2,1,2) model (the order here is only for example), that is ,in , Estimated by historical data , , , Assuming the current order volume Then predict the order volume for the next period:

[0074] (This also requires information based on the residual sequence) Calculations were performed using historical data to simplify assumptions. If all are 0, then (Assuming) The predicted order volume for the next period is... Based on this forecast and the energy consumption characteristics of the stacker crane, the required energy amount is calculated, and the energy supply strategy is adjusted in advance.

[0075] Assuming an energy price threshold is set Yuan / kWh, when the price of traditional electricity is higher than Furthermore, when the energy storage device has sufficient power, it should be used as the primary power source. In the electricity usage decision model, the cost of renewable energy should be considered. Yuan / kWh, traditional electricity cost Yuan / kWh, let the decision variable for using renewable energy be... The decision variables for using traditional electricity are The objective function is to minimize the cost.

[0076] At the same time, it meets energy demand constraints. ( For total energy demand, For renewable energy availability, (The amount that can be provided by traditional electricity). For example, if kilowatt-hours kilowatt-hours For kilowatt-hours, a linear programming model can be established to solve the problem. If the price of renewable energy is low and the electricity supply is sufficient, renewable energy will be used preferentially. It will be quite large.

[0077] Assuming that during the energy switching process, a voltage value sequence before the switching is collected.

[0078] Average voltage value Then the root mean square value of the voltage fluctuation amplitude

[0079] .

[0080] like The value exceeds the set threshold (e.g.) If so, further optimization of the control logic and hardware performance of the hybrid energy switching device is needed, such as adjusting the parameters of the voltage regulator circuit or increasing the capacity of the filter capacitor.

[0081] A goods dispatching and distribution management system suitable for stacker crane automated warehouses includes a personnel operation and training assistance module. This module provides personalized customization of the human-machine interface (HMI), allowing operators to adjust the interface layout, button positions, and function display methods according to their habits and needs. It also establishes a user feedback mechanism to regularly collect operator opinions and suggestions on the interface, and optimizes and improves it based on feedback. Furthermore, it establishes an automatic training content update mechanism. When system functions are updated, the training content update process is automatically triggered, using automated scripting technology to quickly generate training materials for the new functions, including online courses and simulated operation scenarios, and promptly pushes them to operators. Simultaneously, it regularly organizes on-site training activities, inviting system developers to provide detailed explanations and demonstrations of new functions, ensuring that operators can promptly master the latest system features. A security review mechanism for personalized HMI customization is also established. When operators customize the interface, the system automatically performs a security assessment of the customized content, prompting or prohibiting settings that may affect security. Finally, it includes a manual review step added to the automatic training content update mechanism, where professional trainers review and optimize the automatically generated training materials to ensure the accuracy, completeness, and comprehensibility of the training content.

[0082] 1. Human-computer interaction and feedback:

[0083] Assuming an evaluation of the ease of use of the human-computer interaction interface is conducted, 10 operation tasks are recorded, and the operation times are as follows: Seconds, number of operation errors Set a convenience score (Adding 1 here is to avoid the denominator being zero). For example, the time of the first operation. seconds, number of errors ,but Calculate the average convenience score. ,like If the value is low, the button position or function will be redesigned based on operator feedback to improve ease of operation.

[0084] 2. Training System and Audit:

[0085] Assuming operators are assessed after training, what is the accuracy rate of their operation for the new inventory counting process? (i.e., the percentage of correct operations out of the total number of operations), operation speed improvement rate (Assuming the average operation time before training is) seconds, the average operation time after training is seconds, then ), and set a training effectiveness score. Assuming , ,but By monitoring To evaluate the effectiveness of training, if If the value does not meet the expected target, optimize the training content and methods, such as increasing the difficulty of simulated operation scenarios or providing more case studies.

[0086] An inventory dispatching and distribution management system suitable for stacker crane automated warehouses includes an artificial intelligence prediction module. This module incorporates an external data monitoring and fusion mechanism to collect external market data in real time, such as social media trending topics and policy and regulatory changes. This data is then combined with internal business data as input to the prediction model. Simultaneously, an adaptive adjustment mechanism is established to automatically adjust model parameters or structure when significant changes in external data are detected, improving the model's response speed and accuracy to market changes. Furthermore, a visual prediction result display tool is developed, using charts, graphs, and other intuitive methods to present prediction results and key indicator trends. It also provides data interpretation and decision-making suggestions, automatically generating easy-to-understand analysis reports and decision recommendations based on the prediction results, helping enterprise managers quickly understand the prediction information and make reasonable decisions. An external data quality assessment and cleaning system is established to conduct credibility assessments, data cleaning, and format conversion of external data from different sources, ensuring data reliability and consistency. The module also expands the functionality of the visual prediction result display tool, adding multi-dimensional data interactive display and analysis functions, such as using pivot tables and interactive dashboards, allowing enterprise managers to flexibly adjust the display dimensions and analysis perspectives according to their decision-making needs.

[0087] 1. Prediction algorithms and data fusion:

[0088] For the ARIMA model to predict the quantity of goods entering the warehouse, it is assumed that the monthly quantity of goods entering the warehouse over the past two years is as follows: The autoregressive order of the ARIMA model is determined using the AIC criterion. moving average order Difference order After establishing an ARIMA(2,1,2) model and performing parameter estimation and model fitting, the quantity of goods entering the warehouse in the next two weeks is predicted. Assume the predicted quantity of goods entering the warehouse in the first month is... The second month is .

[0089] For the neural network model, assuming the input layer has four nodes representing the month (values ​​1-12), season (spring, summer, autumn, winter, represented by 0-3), the average order volume over the past two months, and whether there are any promotional activities, the output layer is the predicted quantity of goods entering and leaving the warehouse this month. For example, the current month is May (corresponding to an input node value of 5), it is in spring (input node value of 0), and the average order volume over the past two months is... And there have been no recent promotional activities (input node value is 0). Assume the neural network, trained with a weight matrix and threshold vector, calculates the predicted value through forward propagation. Let the input vector... Hidden layer calculation

[0090] ( For activation functions, such as the sigmoid function, the output layer is calculated. This yields the predicted number of items entering the warehouse this month. This is just an example of the calculation framework; the actual weights and thresholds need to be obtained through training with a large amount of data.

[0091] Suppose that mentions of a key beverage on social media increased by [percentage missing] within a week. ,Right now Internal order data shows that the beverage's order volume has recently increased. ,Right now Let the external data influence coefficient be set. Assuming , ,but .according to Adjust the parameters of the prediction model, such as increasing the weight of features related to social media mentions.

[0092] 2. Decision support and visualization:

[0093] Suppose we analyze the work efficiency at different times in the morning and afternoon. From 9-11 AM, [the following data was processed]. There are [number] tasks, with a total transport distance of [distance]. Rice, time spent Hour( (minutes), then the morning work efficiency meters per minute; processed between 2-4 pm There are [number] tasks, with a total transport distance of [distance]. Rice, time spent Hour( (minutes), then afternoon homework efficiency Meters per minute. Based on this result, it is recommended to schedule urgent order processing or complex cargo handling tasks in the morning, and simple inventory organization or equipment maintenance tasks in the afternoon.

[0094] Assuming a credibility assessment is conducted on social media data, the credibility score of the data source is calculated. (Assuming the social media platform has high visibility and credibility), data integrity score (The data includes key information such as mention count and sentiment), data accuracy score. (After preliminary verification, the data showed no obvious errors), assuming a data confidence level. Assuming , , ,but .according to Value filtering and processing of external data, if The value is lower than the set threshold (e.g.) If the data is not properly reviewed, it will be further examined or discarded.

[0095] Suppose a business manager uses a pivot table to view inventory forecasts for different regions, order types, and product categories. They find that the forecast quantity of goods received in region A for product category X and order type Y will increase in the coming week. At the same time, inventory costs will also rise. By dynamically adjusting the time range to the next two weeks through an interactive dashboard and filtering data based on products from specific suppliers, in-depth analysis revealed a strong positive correlation between the quantity of these supplier's products entering and leaving the warehouse and their inventory costs over a certain period (correlation coefficient). Let the improvement rate of decision accuracy be... (Assuming the accuracy of decision-making after data analysis increases from...) Upgraded to ,but Decision speed improvement rate (Assuming the average time taken for previous decisions) Hours, now taking time Hours,

[0096] ), and set a validity score Assuming , ,but By monitoring Value optimization visualization tool functions and display methods, if If the value is low, add more data display dimensions or optimize interactive operations, etc.

[0097] An automated storage and retrieval system (AS / RS) for goods dispatching and distribution management includes an AS / RS and robot collaborative operation module. This module employs a hybrid wired and wireless communication network with a communication redundancy mechanism. When wireless communication is normal, it uses wireless communication for data transmission. When wireless communication is interfered with or interrupted, it automatically switches to a wired backup line to ensure data transmission stability. Simultaneously, it optimizes the wireless communication protocol and signal enhancement technology to reduce the impact of the metallic environment on the signal. Furthermore, a unified robot task allocation and traffic management system is established to centrally schedule and plan the tasks of AGVs and shuttles. During path planning, algorithms such as time window allocation and priority queuing are used to avoid robots getting stuck in aisles and intersections. To address conflicts, such as allocating specific time windows for each robot to enter narrow passages and prioritizing tasks based on their urgency, the lower-priority robot actively avoids or waits when paths conflict with those of a higher-priority robot. This also includes optimizing the switching protocols and hardware for wired and wireless hybrid communication networks, employing double-buffering technology and a seamless switching mechanism. Before switching, data to be transmitted is temporarily stored in a buffer area, and the buffered data is transmitted immediately after switching to ensure no data loss. Furthermore, a dynamic priority adjustment and emergency handling module is introduced into the robot task allocation and traffic management system. When unexpected tasks or equipment failures occur, the robot's task priority and path planning are dynamically adjusted based on factors such as task urgency and remaining equipment resources, prioritizing urgent tasks.

[0098] 1. Communication and Task Assignment:

[0099] Assuming a total of [number] wireless communication transmissions Number of successful transmissions Then the reliability of communication .like The value is lower than the set threshold (e.g.) If necessary, optimize the wireless communication protocol, such as adjusting the transmission power, optimizing the data packet size, or changing the communication frequency band. At the same time, check the effect of signal enhancement technology, such as whether the antenna array layout is reasonable.

[0100] 2. Traffic Management and Emergency Response:

[0101] Assuming a certain channel's busyness coefficient (Calculated based on the number of robots passing through the channel or the workload per unit time), robot operation efficiency (Calculated based on the amount of tasks completed by the robot per unit time), initial time window Minutes, set time window Assuming , ,but minutes, that is, the adjusted time window is Minutes are allocated to improve channel utilization.

[0102] Assuming a total number of data transmissions Number of data transmission interruptions Then the data transmission continuity score .like If the value is low, optimize the buffer size of the double buffering technology or the response speed of the switching mechanism to ensure the continuity of data transmission and avoid task interruption or data errors caused by switching.

[0103] Assuming the urgency level of a certain urgent task Remaining equipment resources (For example, the remaining battery power is) The remaining load capacity is (etc.), let the task priority adjustment formula be: Assuming initial priority , , ,but (Priority can be normalized according to actual conditions). If this task's priority is increased, the relevant robot will handle it first and coordinate with other robots to give way. Assume the emergency task completion time... Minutes, normal task completion time minutes, then the emergency response efficiency By monitoring Values ​​are used to assess the effectiveness of emergency response strategies. The value did not meet the expected target (e.g., less than). If so, the emergency resource allocation plan should be adjusted or optimized.

[0104] For example, if a frequently used path is found to be congested, resulting in low efficiency in emergency response, the weight of alternative paths can be increased or the overall path strategy can be replanned to guide the robot to select smoother paths to perform tasks.

[0105] Suppose that during an emergency equipment failure response, a stacker crane malfunctions, and the AGV tasks originally awaiting service from that crane are reassigned. Assume the original estimated completion time for the AGV tasks is... After minutes, the task was re-planned and assigned to a new stacker crane to complete, with the actual time taken being [time missing]. minutes. This indicates a time saving rate for this emergency response. This indicates that the emergency response strategy saved time in completing the task. Statistical analysis of multiple emergency response situations If the average value is low, further analysis should be conducted to determine the reasons, such as whether the robot scheduling algorithm is not flexible enough or communication delays cause untimely information transmission, and targeted improvements and optimizations should be made.

[0106] Regarding the collision situation of the robot operating within the channel, assuming the number of potential collisions detected within one hour is... After adjustments using time window allocation and priority queuing algorithms, the potential number of conflicts in the next hour was reduced to [number missing]. Conflict reduction rate That is, the conflict has decreased. .like The value did not reach the ideal level (e.g., less than) This would further refine the time window allocation strategy. For example, based on factors such as differences in robot speed and the urgency of task types, the allowed passage time window for each robot in the channel could be dynamically adjusted. Alternatively, the priority judgment factors in the priority queuing algorithm could be optimized, taking into account not only the urgency of the task but also the robot's remaining battery power and distance from the target point, in order to more accurately avoid conflicts and improve the smoothness and efficiency of warehouse operations.

[0107] As shown above, the intelligent scheduling module employs a dynamic algorithm switching mechanism combined with a task transition algorithm. When faced with high order volumes and complex tasks, it can quickly switch algorithms and smoothly transition, avoiding task delays and confusion that may occur during algorithm switching in traditional scheduling systems. The special cargo database and its real-time updates, along with the formulation of targeted scheduling rules, provide refined management for the specific needs of special cargoes. This differs from the general scheduling system's uniform approach to regular cargo, improving the accuracy and security of special cargo scheduling.

[0108] The precision inventory management module: The environment adaptation module in the data fusion algorithm adopts regionally differentiated threshold settings and judgment logic, fully considering the environmental characteristics of different areas of the warehouse. This enables more accurate processing of sensor data and reduces inventory data errors caused by environmental interference. The AI-assisted manual review mechanism uses machine learning models to pre-screen and classify goods, improving review efficiency and accuracy, a stark contrast to traditional inventory management systems that rely solely on manual or simple automated reviews.

[0109] System Integration Module: The innovations of this solution lie in its node fault isolation and self-healing mechanism for the middleware cluster and its dynamic monitoring platform for multi-system interface version compatibility. During system integration, it proactively addresses middleware node failures, ensuring data transmission stability, while simultaneously monitoring interface version compatibility in real time to promptly identify and resolve potential problems. This is a relatively advanced feature in existing system integration technologies, improving the reliability and scalability of system integration.

[0110] The equipment failure response module features an innovative distributed computing architecture that accelerates data processing for incremental learning algorithms and optimizes real-time monitoring information processing (event-driven architecture and intelligent caching technology). It can quickly update the failure prediction model and adjust scheduling schemes promptly, effectively reducing job interruptions and delays caused by equipment failures. This improves the system's ability to respond to equipment failures and its overall operational efficiency, demonstrating significant advantages over traditional equipment failure response systems.

[0111] Energy Management Module: The real-time order correction mechanism complements the energy-efficiency balance model and features innovative improvements in the optimized control logic and hardware performance of the hybrid energy switching device. By dynamically correcting energy allocation strategies based on real-time order information, energy management becomes more precise and flexible. The optimized hybrid energy switching device ensures a smooth transition during energy switching, improving energy utilization efficiency and system stability, representing a significant improvement over existing energy management technologies.

[0112] The personnel operation and training assistance module features innovative aspects, including a personalized human-computer interaction interface, a security audit mechanism, and a manual review component within the automatic training content update mechanism. It provides an effective solution for ensuring operational safety and improving training quality. Compared to general personnel operation and training assistance systems, it places greater emphasis on the individualized needs of operators and the accuracy of training content, helping to reduce human error and improve the overall system performance.

[0113] The AI ​​prediction module distinguishes itself from other AI prediction technologies by its multi-dimensional data interaction capabilities, including an external data quality assessment and cleaning system and expanded visualization tools for predictive results. Through reliable processing of external data and multi-dimensional data interaction, it provides enterprise managers with more comprehensive and in-depth decision support. Its innovative approach to data fusion and visualization better meets the needs of complex decision-making.

[0114] The automated storage and retrieval system (AS / RS) and robot collaborative operation module is unique in its dual-buffering technology and seamless switching mechanism using a hybrid wired and wireless communication network, as well as its dynamic priority adjustment and emergency handling modules within the robot task allocation and traffic management system. This effectively ensures stable communication and responds to emergencies, improving the efficiency and reliability of AS / RS and robot collaborative operations. Compared to general AS / RS collaborative operation technologies, it represents a significant technological advancement in communication and task management.

[0115] In summary, this technical solution achieves efficient, intelligent, and reliable operation of the stacker crane automated warehouse goods dispatching and distribution management system through the close collaboration of multiple modules and the comprehensive application of innovative technologies. Compared with existing technologies, it does not simply piece together individual functional modules, but rather optimizes each aspect of the system as a whole, forming an organic whole. For example, the dynamic algorithm switching mechanism of the intelligent dispatching module, the environmental adaptive data fusion of the precise inventory management module, and the interface compatibility management of the system integration module work together to quickly adapt to order changes, environmental interference, and system integration requirements in complex and ever-changing warehouse operating environments, thereby improving the overall adaptability and stability of the system.

[0116] The data sharing and interaction mechanisms between modules also reflect technological innovation. The unified data platform built through the system integration module enables real-time and accurate data transmission between different modules, allowing each module to make more rational decisions based on global information. For example, the prediction results from the artificial intelligence prediction module can be promptly transmitted to multiple modules such as intelligent scheduling and energy management, enabling these modules to prepare in advance. This cross-module data-driven decision-making mechanism is rare in existing technologies.

Claims

1. A goods dispatching and distribution management system suitable for stacker crane automated warehouses, characterized in that: include: The intelligent scheduling module employs a multi-algorithm fusion and dynamic switching mechanism to plan stacker crane paths and allocate tasks based on orders and warehouse conditions. It also includes a special cargo handling strategy. The multi-algorithm fusion of the intelligent scheduling module includes a combination of A* or Dijkstra's algorithm and task allocation algorithms. The dynamic switching mechanism switches between conventional and fast scheduling algorithms based on order volume and computing resource availability. The special cargo handling strategy includes a special cargo database with its real-time updates and the formulation of targeted scheduling rules. Precision Inventory Management Module: Utilizes multi-sensor fusion and intelligent recognition technology, combined with environmental adaptation and intelligent auditing to ensure accurate inventory data; System integration module: It realizes multi-system integration through standardized interfaces and middleware, and has cluster deployment, fault handling and interface management mechanisms; Equipment Failure Response Module: This module utilizes multiple sensors and machine learning algorithms to achieve fault detection and early warning, and leverages distributed computing and real-time monitoring to optimize emergency dispatch. The machine learning algorithm in this module employs a hybrid intelligent dispatch algorithm that combines genetic algorithms and A* algorithms. The distributed computing architecture accelerates data processing for the incremental learning algorithm, and real-time monitoring uses an event-driven architecture and intelligent caching technology to optimize the generation of dispatch schemes. Energy management module: Optimizes energy utilization based on an energy-efficiency balance model combined with a hybrid energy intelligent switching system; Personnel operation and training support module: provides a personalized safety customization interface and a high-quality training system; Artificial intelligence prediction module: Integrates multi-source data and uses visual, multi-dimensional interactive tools to assist decision-making; Automated warehouse and robot collaborative operation module: adopts redundant communication network and dynamic task management strategy to ensure high efficiency of collaboration.

2. The cargo dispatching and distribution management system for stacker crane automated warehouses as described in claim 1, characterized in that: The precision inventory management module uses Kalman filtering or fuzzy logic algorithms to process RFID and laser scanning sensor data through multi-sensor fusion. It also features environmental adaptation to adjust data fusion weights based on regional differences, and intelligent auditing combines artificial intelligence pre-screening with manual auditing.

3. A cargo dispatching and distribution management system suitable for stacker crane automated warehouses as described in claim 1, characterized in that: The middleware of the system integration module adopts Enterprise Service Bus, and the cluster deployment has node fault isolation and self-healing functions. The interface management includes a version compatibility mechanism and a dynamic monitoring platform for multi-system compatibility.

4. A cargo dispatching and distribution management system suitable for stacker crane automated warehouses as described in claim 1, characterized in that: The energy management module's energy-efficiency balance model, combined with a real-time order correction mechanism, adjusts the stacker crane's energy strategy. The hybrid energy intelligent switching system includes voltage stabilization filtering and pre-switching buffer functions.

5. A cargo dispatching and distribution management system suitable for stacker crane automated warehouses as described in claim 1, characterized in that: The personalized security customization interface of the personnel operation and training assistance module is equipped with a security audit mechanism, and the automatic update mechanism of the training system includes a manual audit step to ensure training quality.

6. A cargo dispatching and distribution management system suitable for stacker crane automated warehouses as described in claim 1, characterized in that: The multi-source data fusion of the artificial intelligence prediction module includes an external data quality assessment and cleaning system, and the visualization multi-dimensional interactive tool uses pivot tables and interactive dashboards to achieve multi-dimensional analysis and display.

7. A goods dispatching and distribution management system suitable for stacker crane automated warehouses as described in claim 1, characterized in that: The redundant communication network between the automated warehouse and the robot collaborative operation module adopts double buffering technology and seamless switching mechanism to ensure data transmission. The dynamic task management strategy adjusts the robot task priority and path planning in real time according to the urgency of the task and the equipment status.

Images