Information management method, device, electronic equipment and storage medium for intelligent warehousing

Abstract

The present application relates to the field of warehouse management technology, and provides an information management method, device, electronic device and storage medium for intelligent warehousing. By preprocessing the target data set to realize a standardized spatiotemporal fusion data stream, a virtual space is dynamically mapped based on the standardized data stream to generate a virtual-real synchronized digital twin warehouse model, and multi-target scheduling is performed according to the real-time status data in the digital twin warehouse model to generate an AGV path matrix, and the AGV path matrix is ​​dynamically replanned according to the real-time status data to generate anti-disturbance scheduling instructions. At the same time, root cause analysis of the product data set is performed to obtain predicted improvement data, and finally, material allocation analysis is performed on the anti-disturbance scheduling instructions and predicted improvement data according to the preset optimization method to obtain warehousing and logistics decisions. This application realizes flexible scheduling of warehousing and logistics based on the current status of warehousing and combined with the predictive analysis of the product cycle, which can accurately respond to actual needs and adjust the operation strategy in time.

Description

Technical Field

[0001] The present application relates to the field of warehouse management technology, and in particular to an information management method, device, electronic equipment and storage medium for intelligent warehouse. Background Art

[0002] In order to achieve the reasonable allocation of enterprise storage resources, smart warehouses based on the Internet of Things have emerged, which use sensors and radio frequency identification technology to accurately locate and track inventory goods, thereby reducing inventory data errors and goods loss. Smart warehousing technology based on the Internet of Things can greatly improve the efficiency of warehouse storage links including but not limited to storage, handling, picking and distribution of goods. At the same time, through intelligent scheduling and resource optimization, limited warehouse resources can be used more effectively.

[0003] Although the existing intelligent warehouse management system has improved the operational efficiency of each link, the core algorithm of intelligent scheduling lacks highly integrated collaborative scheduling and data linkage, which makes the real-time data of each link form information islands. It is unable to make optimized decisions on the collaborative scheduling of production materials according to the product cycle changes (i.e., production demand and material dynamic changes, etc.). Summary of the invention

[0004] In view of this, the present application provides an information management method, device, electronic device and storage medium for intelligent warehousing to solve the problem of intelligent warehousing collaborative scheduling based on data linkage.

[0005] The first aspect of the present application provides an information management method for intelligent warehousing, the method comprising: Perform data preprocessing on the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream; Performing dynamic mapping of virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model; Performing multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix; Dynamically replanning the AGV path matrix according to the real-time status data of the AGV device to obtain anti-disturbance scheduling instructions; Acquire a product data set of a target system and perform root cause analysis on the product data set to obtain prediction improvement data; The anti-disturbance scheduling instructions and the predicted improvement data are analyzed for material allocation according to a preset optimization method to obtain real-time warehousing logistics decisions.

[0006] In an optional embodiment, the preprocessing of the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream includes: Collect multi-source heterogeneous data streams in real time to obtain the target data set; Performing time series synchronization processing on the target data set to obtain a time series data set; Performing spatiotemporal fusion processing on the time series data set to obtain a spatiotemporal fusion data stream; Performing data cleaning and interpolation on the spatiotemporal fusion data stream to obtain an advanced spatiotemporal fusion data stream; The advanced spatiotemporal fusion data stream is formatted to obtain the standardized spatiotemporal fusion data stream.

[0007] In an optional embodiment, the step of dynamically mapping the virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model includes: Step S21, performing time synchronization and position matching on the real-time sensor data in the standardized spatiotemporal fusion data stream to obtain first synchronized data; Step S22: constructing a virtual space model according to the first synchronization data and a preset warehouse structure data set to obtain a primary space model; Step S23, performing layered processing on the primary space model to obtain an intermediate space model; Step S24: Mapping the real-time status data and environmental data in the standardized spatiotemporal fusion data stream to the intermediate space model to obtain a digital twin warehouse model at the current moment; Step S24 is repeatedly executed according to the standardized spatiotemporal fusion data stream updated in real time to update the digital twin warehouse model in real time.

[0008] In an optional embodiment, performing multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix includes: Prioritizing each warehousing task point in the digital twin warehousing model to obtain a task execution list; Generating a plurality of potential path sets according to a preset constraint optimization model and the task execution list; Obtain the optimal path corresponding to each storage task point according to the preset multi-objective optimization trade-off algorithm to obtain the optimal path combination; Tasks are assigned to each AGV according to the optimal path combination and the task execution list to obtain the AGV path matrix.

[0009] In an optional embodiment, the dynamically replanning the AGV path matrix according to the real-time status data of the AGV device to obtain the anti-disturbance scheduling instruction includes: Acquire real-time status data of each AGV device according to the AGV path matrix to perform abnormality detection according to the real-time status data; Dynamically adjust the AGV path matrix according to a preset model control prediction algorithm and detected abnormal information; The anti-disturbance scheduling instruction is generated according to the adjusted AGV path matrix and the task execution list.

[0010] In an optional embodiment, acquiring a product data set of a target system and performing root cause analysis on the product data set to obtain prediction improvement data includes: Acquire a product data set of a target system and identify production requirements according to the product data set to obtain first analysis data; Performing capacity analysis according to the product data set to obtain second analysis data; Performing matching analysis on the inventory material data in the target data set according to the first analysis data to obtain inventory forecast fluctuation data; Perform product cycle analysis according to the product data set to obtain third analysis data; Performing root cause analysis on the first analysis data, the second analysis data, the inventory forecast fluctuation data, and the third analysis data according to a preset Bayesian network algorithm to identify potential influencing factors; An improved derivation is performed on each influencing factor to obtain the prediction improvement data.

[0011] In an optional embodiment, performing material allocation analysis on the anti-disturbance scheduling instructions and the prediction improvement data according to a preset optimization method to obtain real-time warehousing logistics decisions includes: Prioritizing the anti-disturbance scheduling instructions to obtain a material scheduling matrix; Performing demand fluctuation analysis on the forecast improvement data to obtain a demand forecast curve; Comparing the demand forecast curve with the inventory material data to obtain inventory warning data; Generate a material dispatch instruction according to the anti-disturbance dispatch instruction and the inventory warning data; Generate a path optimization instruction according to the material dispatch instruction and the material dispatch matrix; The task execution list in the digital twin warehousing model is feedback-scheduled according to the material scheduling instructions and the path optimization instructions to obtain the real-time warehousing logistics decision.

[0012] The second aspect of the present application provides an information management device for intelligent warehousing, the device comprising: A data fusion module is used to perform data preprocessing on the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream; A model building module, used for dynamically mapping virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model; A path matrix module is used to perform multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix; An instruction adjustment module is used to dynamically replan the AGV path matrix according to the real-time status data of the AGV equipment to obtain anti-disturbance scheduling instructions; An improvement prediction module, used for acquiring a product data set of a target system and performing root cause analysis on the product data set to obtain prediction improvement data; The logistics decision module is used to perform material allocation analysis on the anti-disturbance scheduling instructions and the predicted improvement data according to a preset optimization method to obtain real-time warehousing logistics decisions.

[0013] A third aspect of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the information management method for intelligent warehousing as described above when executing the computer program.

[0014] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the information management method for intelligent warehousing as described above.

[0015] In summary, this application at least includes the following beneficial technical effects: 1. Real-time mapping of the storage environment and operation process is achieved by building a virtual-real synchronized digital twin storage model. The real-time perception of changes in the storage environment is improved, and the storage status can be monitored in real time through dynamic updates during system operation to ensure that the system can quickly respond to changes in the environment (for example, displacement of items, equipment failure, etc.), thereby effectively improving the flexibility of storage management.

[0016] 2. Through in-depth mining of product data sets through root cause analysis, potential factors that may affect warehouse efficiency (for example, production demand fluctuations, inventory management issues, etc.) can be identified, and predictive improvements can be made based on the analysis results.

[0017] 3. Based on the prediction improvement data and anti-disturbance dispatch instructions, the system can generate accurate material allocation instructions and combine inventory warning data for inventory management. It helps to replenish and reasonably dispatch storage materials in time, avoids the risk of material waste and inventory shortage, and can further improve logistics efficiency through path optimization instructions, making material allocation more intelligent and automated. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flow chart of an information management method for intelligent warehousing provided in an embodiment of the present application; Figure 2 It is a functional module diagram of an information management device for intelligent warehousing provided in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] like Figure 1 The figure is a flow chart of the information management method of the intelligent warehousing provided by the embodiment of the present application. The information management method of the intelligent warehousing provided by the embodiment of the present application includes the following steps.

[0022] Step S1: preprocess the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream.

[0023] It should be understood that in the intelligent warehousing system in the embodiment of the present application, various sensors, AGV positioning modules, RFID tags, environmental detection equipment and other IoT devices are deployed. Among them, RFID tags are used to track the inbound and outbound status of goods in real time and provide unique identification information of goods; AGV positioning modules are used to obtain the real-time location and driving trajectory of AGVs through laser radar, GPS and ultrasonic sensors; sensors are used to monitor the environmental conditions of the warehouse and the operating status of equipment and AGVs.

[0024] The monitoring data is transmitted to the cloud processing platform in real time through wireless communication protocols (such as Wi-Fi, Zigbee, and Lora). The cloud processing platform organizes these data streams into target data sets and stores them uniformly. Since the data sources (i.e., various data acquisition sensors) are usually heterogeneous and use different formats, acquisition frequencies, and accuracies, the multi-source heterogeneous data streams collected in real time in the target data set need to be processed uniformly.

[0025] For the multi-source heterogeneous data streams in the target data set, the following formula is used to perform time series synchronization processing: .

[0026] in, is the ith data stream after synchronization, is the i-th data item in the original data stream, is the standard time base, It is a mapping function for timing synchronization. Through timing synchronization, all data are aligned according to the same time window, eliminating the time misalignment problem caused by inconsistent device acquisition frequency, thereby uniformly converting the timestamps in each data stream, and merging different data streams based on the preset time standard to obtain a time series data set.

[0027] For time series data sets, different types of data (for example, environmental data, inventory data, equipment data, etc.) need to be fused in time and space. Through time and space fusion, data from different sources distributed in time and space are integrated together to form a complete data stream. For time series data sets, the following formula is used for time and space fusion processing: .

[0028] in, is the fused data stream (i.e., spatiotemporal fused data stream), is the i-th time series data stream (e.g., environmental data, inventory data, etc.), is the weight of the ith data stream, is the number of data streams. Through multiple sensor data fusion algorithms (e.g., Kalman filter, particle filter, etc.), the temporal and spatial information of each data source is comprehensively processed through spatial positioning and time synchronization. Specifically, the inventory location is spatially connected with the AGV location, the environmental data is combined with the equipment status, and a unified data stream (i.e., spatiotemporal fusion data stream) is generated through multi-dimensional analysis.

[0029] In the processing of spatiotemporal fusion data streams, data cleaning and interpolation is a key step, the goal of which is to remove outliers, fill in missing data, and ensure data integrity and continuity. Outliers are usually caused by sensor errors, environmental interference, or equipment failure. In the cleaning step, we need to use an outlier detection algorithm to identify these data that do not conform to the expected rules and eliminate them.

[0030] For each data point, the following Z-score formula is used for outlier detection: in, is the Z-score value of the ith data point, is the original value of the ith data point, is the mean of the data set, is the standard deviation of the data set. The Z-score is used to calculate the degree of deviation of the data. If the absolute value of the Z-score is greater than the preset threshold, the data point is considered to be an outlier and needs to be processed or eliminated.

[0031] For the cleaned data, if there is data missing in certain time periods or spatial locations, the system will interpolate and supplement it based on the surrounding data. For example, during the AGV path planning process, if the location information of a certain section of the path is missing, the data of the surrounding time points can be interpolated to ensure the continuity of the path planning.

[0032] For missing data due to equipment failure or environmental factors, interpolation methods are needed to fill in the missing values. Common interpolation methods include linear interpolation, spline interpolation, Kalman filtering, etc. In order to maintain smoothness and continuity in the data stream, the embodiments of the present application generally select linear interpolation and spline interpolation as common interpolation methods.

[0033] For missing data between adjacent acquisition time points, the following linear interpolation formula is used to compensate for missing values: in, is the interpolation result at time t, For time point The known data at For time point The known data at is the timestamp of the known data point, and t is the time point to be interpolated.

[0034] For some spatiotemporal data with large fluctuations, linear interpolation may not accurately reflect the changing trend of the data. In this case, spline interpolation can be used. Spline interpolation is to fit the data points through a smooth polynomial curve. This embodiment performs spline interpolation through the following formula: in, is the i-th segment spline interpolation curve, is the time of the known data point, , , , are the interpolation coefficients calculated from known data points.

[0035] By cleaning and interpolating data, invalid data caused by acquisition equipment failure, environmental interference and other factors can be effectively removed, thereby obtaining accurate and reliable advanced spatiotemporal fusion data streams.

[0036] In order to meet the format requirements of subsequent analysis, the cleaned and interpolated advanced spatiotemporal fusion data stream needs to be converted into a unified format that meets the needs of subsequent analysis. By formatting, the data will be converted into a structured table or database form so that it can adapt to the input requirements of different analysis modules. Formatting also includes the integration of different data dimensions, such as uniformly encoding inventory data, environmental data, and equipment data according to the spatiotemporal dimension to ensure that the system can read and process it smoothly. By formatting the advanced spatiotemporal fusion data stream, a standardized spatiotemporal fusion data stream is obtained to ensure that the subsequent decision analysis module can smoothly use this data for further analysis and decision support.

[0037] Step S2: Dynamically map the virtual space according to the standardized space-time fusion data stream to obtain a digital twin warehouse model that is synchronized between the virtual and the real.

[0038] Since different sensors have certain time delays and deviations when collecting data, it is necessary to further perform time synchronization processing on the basis of timing synchronization. Each sensor data stream has a timestamp, so by uniformly correcting the timestamps of all sensors, data from different sources can correspond to the same time. For example, the data of temperature and humidity sensors and AGV position sensors may have slight time differences, and unifying these time differences ensures the relevance of the data. Use the following formula for time synchronization: in, is the calibrated timestamp, is the original timestamp of the ith sensor, To unify the reference time.

[0039] After completing the time synchronization of real-time sensor data, the sensor position is calibrated to map the data collected by the sensor to the actual physical location. With the known warehouse floor plan and the geographic coordinates of each sensor, the coordinate conversion algorithm is used to correct the position of each sensor to the corresponding position in the virtual warehouse space. The following formula is used to match the position of each sensor: in, is the location data after matching, is the original sensor position data, The conversion function defined by the coordinate transformation matrix A is used to convert the actual position data into coordinates in the virtual space.

[0040] The first synchronization data is obtained by time synchronization and position matching of real-time sensor data. The first synchronization data includes but is not limited to the synchronized timestamp and the matched location information, which provides basic data support for the virtual warehousing model. Combined with the first synchronization data and the preset warehouse structure data set (for example, shelf location, channel layout, equipment distribution, etc.), by establishing a three-dimensional model of the virtual space, a primary virtual warehouse model (i.e., a primary space model) is further constructed. The primary space model includes but is not limited to the basic space layout of each area in the warehouse, such as the arrangement of shelves, the storage location of goods, the division of channels, and other information.

[0041] Using the preset warehouse structure data, a spatial structure consistent with the actual warehouse is created in the virtual space through a 3D modeling tool (e.g., Unity3D engine). Each physical element (e.g., shelf, AGV path, equipment location, etc.) has a clear 3D coordinate and spatial position in the virtual model. The synchronized first synchronization data and warehouse structure data are integrated to determine the corresponding coordinates of each sensor position in the virtual space.

[0042] Through the preset model layering strategy, different types of storage elements are classified into different levels. For example, the shelf layer contains the status of all shelves and goods, the equipment layer contains all AGV, sensor and other equipment information, and the environment layer contains the warehouse's temperature, humidity, lighting and other environmental data. According to the specific layout and functional requirements of the warehouse, the spatial information of each level is integrated into the intermediate space model one by one. Each layer can be updated independently, and the layers are consistent through data mapping to ensure that the data transmission between each level is correct.

[0043] According to the data in the real-time sensor data stream (for example, AGV location, temperature and humidity, shelf status, etc.), the data of each level in the intermediate space model is updated to obtain the digital twin warehouse model corresponding to the current moment. For example, the inventory status of the shelf, the location and task status of the AGV, etc. are updated in real time.

[0044] In order to ensure that the virtual warehouse model always maintains consistency with the actual warehouse, the real-time update data stream will continuously trigger the update process of the virtual warehouse model. That is, every time the sensor collects new data, each frame of data in the standardized spatiotemporal fusion data stream is continuously mapped and updated to the virtual warehouse model. The time, location, environment and other information in each data packet will be fed back to the virtual model in real time to ensure that the status of the digital twin warehouse model is synchronized with the actual warehouse.

[0045] Step S3: Perform multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix.

[0046] It should be understood that since the intelligent warehousing system of the embodiment of the present application will receive logistics dispatch instructions from different platforms in real time, the logistics dispatch instructions received at the current moment will be marked together with the digital twin warehousing model when the digital twin warehousing model is being constructed, so as to facilitate display to management personnel. Among them, the logistics dispatch instructions include but are not limited to task points (for example, storage, pickup, and transshipment of goods), execution time limits, and urgency.

[0047] The priority of the current task itself is determined based on the time limit and urgency when the logistics dispatch instruction is received (for example, the execution of customer orders may need to take precedence over inventory replenishment tasks), and the occupation of storage space when the storage task is executed is obtained based on environmental sensors. At the same time, each task point is scored using a weighted scoring method in combination with the available status of the AGV and related equipment at the current moment. Specifically, each task is assigned a weight based on different factors (such as urgency, space occupancy, equipment status, etc.), and the final task priority is calculated. The task priority list will be arranged in order of priority to form a task execution list.

[0048] Therefore, based on the time sequence and spatial relationship between tasks in the task execution list, multiple potential path sets are selected using a preset constraint optimization model. The constraint optimization model is used to minimize the total cost of the path between task points (such as time, energy consumption, etc.) and consider multiple constraints, such as path capacity, equipment load, time window, etc. The constraint optimization model can be expressed as the following formula: in, Describes the path cost from task point i to task point j. The path cost can be time, energy consumption, distance, etc. is a binary decision variable used to indicate whether to select path i to j. When the path is selected =1, otherwise =0. is the total number of task points, indicating the number of tasks that need to be scheduled. is the number of path selections, which indicates the number of optional paths from task point i to task j. The optimal path that minimizes the total cost is selected through the constraint optimization model, so that the sum of the path costs of each task point is minimized while satisfying the constraints (such as task time window, AGV capacity, etc.). Finally, the constraint optimization model provides feasible path selections between each task point and forms a potential path set.

[0049] After obtaining multiple potential path sets, the weighted sum method is used to balance different optimization objectives (such as path length, energy consumption, task execution time, etc.), and a combined objective function is generated for optimization. Through the weighted sum method, multiple optimization objectives can be integrated so that they work together on path selection. The objective function of multi-objective optimization is expressed as the following formula using the weighted sum method: in, is the kth objective function, which represents the value of the kth optimization goal. (For example, the first goal may be the path length (shortest path), the second goal may be the energy consumption, the third goal may be the task execution time, etc.) is the weight of the kth objective function, indicating the relative importance of each objective in the overall optimization. The sum of all weights should be 1. is the total number of objective functions, that is, the number of objectives involved in optimization. The importance of different objectives is balanced by adjusting the weight of each objective function, so that the final path selection can find a suitable compromise between multiple optimization objectives. During the optimization process, by solving the objective function, the optimal path combination corresponding to each task point is obtained. These path combinations comprehensively consider multiple factors such as path length, energy consumption, time, etc., and provide the optimal path planning result (that is, the optimal path combination).

[0050] After obtaining the optimal path combination for each task point, these tasks are assigned to the AGV in combination with the task priority information in the task execution list. Task allocation needs to consider multiple factors, including the current status of the AGV (such as charging status, availability), path selection and task priority, path load, etc. According to the priority of the task and the status of the AGV, a task allocation algorithm (such as shortest task priority, load balancing, etc.) is used to reasonably allocate tasks to ensure that the task volume of each AGV is balanced and the path selection is optimized. The paths of all AGVs are comprehensively considered to ensure that the tasks are completed while avoiding path conflicts. Through task allocation, an AGV path matrix containing task execution order and path selection is finally formed.

[0051] Step S4: dynamically replan the AGV path matrix according to the real-time status data of the AGV equipment to obtain anti-disturbance scheduling instructions.

[0052] Obtain real-time status data from each AGV device in the warehouse environment. This data comes from a variety of sensor devices, such as: position sensors (used to obtain the position coordinates of the AGV), speed sensors (used to obtain the driving speed of the AGV), load sensors (used to detect the current load of the AGV), vibration sensors (used to monitor the health of the equipment operation), etc. Perform anomaly detection on the corresponding AGV equipment based on the real-time status data to identify faults or interference factors that may affect the execution of the path. For example, if the load of the AGV exceeds the predetermined range, the operating efficiency of the equipment may be reduced; if the vibration sensor feedback of the AGV shows that the equipment is faulty, it may need to be temporarily shut down or maintained; if certain paths in the warehouse are blocked, the path of the AGV will be affected.

[0053] In the embodiment of the present application, anomaly detection is performed on the AGV equipment through a model based on machine learning, and the collected real-time status data is analyzed to check whether the following situations exist: 1. The AGV operation status deviates from the normal range (for example, abnormal speed, excessive vibration, blocked path); 2. The environment changes suddenly (for example, a shelf is moved or the AGV’s driving path is blocked by an obstacle).

[0054] When an abnormal device is detected, the preset model prediction control (MPC) algorithm is used to dynamically adjust the path of the AGV. The MPC method optimizes by considering the state, control variables, and future state of the system to avoid future deviations and path interruptions. Specifically, MPC calculates a set of possible path trajectories based on real-time state data (such as the position, speed, load, etc. of the AGV) and the current environmental state of the warehouse (such as path blockage, obstacles, etc.), and selects an optimal trajectory for adjustment. The core idea of ​​MPC is to predict future states within a given time window and adjust the control input (motion trajectory of the AGV) to minimize the objective function. The minimization of the objective function is expressed using the following formula: in, is the objective function, which represents the cost of path optimization. is the position of the AGV at time t. is the target location. is the control input (e.g., the speed and direction of the AGV). Q and R are weight matrices used to balance the cost of path deviation and control input. By minimizing the objective function, the MPC optimization algorithm can select the most appropriate path and adjust the movement of the AGV in a dynamically changing warehouse environment to avoid collisions and reduce energy consumption.

[0055] After the path adjustment, the next step is to consider the dynamic adjustment of task priorities. In warehouse management, different tasks may have different urgency. For example, some AGVs may be performing high-priority cargo handling tasks, while other AGVs may only be performing storage tasks. Therefore, it is necessary to re-evaluate the task priority and adjust the path execution order in combination with the task execution list. Based on the status of the AGV (e.g., load, task urgency, etc.) and the results of the path adjustment, combined with the order of tasks in the warehouse, the task arrangement of the AGV is re-determined through the optimization scheduling algorithm.

[0056] The final anti-disturbance scheduling instructions are generated based on the adjusted AGV path matrix and task execution list. Anti-disturbance scheduling instructions include but are not limited to AGV path adjustment information, task execution order, scheduled time window, path obstacle avoidance and other necessary instructions. To achieve visual management, anti-disturbance scheduling instructions will be displayed in the warehouse management system and transmitted to the AGV control system for real-time execution.

[0057] Step S5: Acquire a product data set of the target system and perform root cause analysis on the product data set to obtain prediction improvement data.

[0058] It should be understood that the intelligent warehousing system of the present application is an information integration platform integrating multiple systems, including but not limited to Enterprise Resources Planning (ERP) and Product Lifecycle Management (PLM).

[0059] By accessing ERP data (including customer order information, product sales plans, and product demand forecasts) in multiple integrated systems (i.e., target systems), current and future production needs are identified. By extracting information such as order quantity, product specifications, delivery dates, and customer priorities from product data sets. Combining customer orders with historical sales data, actual production needs are calculated to obtain a production demand list. Demand is identified for each product category, including product quantity requirements, production cycles, order urgency, etc., and by comparing with inventory data, the demand gap and production pressure for each product are determined. Thus, the first analysis data (i.e., specific production requirements for each product) is obtained, which includes but is not limited to the quantity required to be produced, production priority, and the gap with existing inventory.

[0060] According to the production process, production line capacity, material consumption and production scheduling information in the target system, the maximum production capacity of each production line can be estimated. Combined with the first analysis data, a capacity assessment is conducted to understand whether the existing production capacity can complete the production task on time or whether the capacity needs to be increased. Capacity analysis is not limited to the evaluation of production lines, but also includes the evaluation of production processes. If there are bottlenecks in certain production links (such as a particular process or equipment is too crowded), these bottleneck information needs to be included in the capacity analysis results to further deduce the potential bottlenecks of product production and the risk of production delays. In this way, the second analysis data (i.e., detailed analysis data on production capacity) is obtained. The first analysis data includes but is not limited to information such as production bottlenecks, equipment load conditions, and available resources for production lines.

[0061] The production demand information in the first analysis data is matched with the inventory data in the target data set for analysis. This is to evaluate whether the existing inventory can meet the production demand and predict inventory fluctuations. Specifically, first, the inventory data (for example, the existing inventory quantity, arrival time, usage rate, etc. of each material) is compared with the production demand to calculate the shortage or surplus of each material. Based on historical inventory fluctuation data, material procurement cycle, supplier delivery status and other factors, the trend of inventory level changes is predicted to obtain inventory fluctuation data. Matching analysis reveals the possible risks of inventory in the future, such as excess or shortage of materials, and provides a basis for warehousing and production scheduling. Thus, inventory forecast fluctuation data (that is, detailed analysis results on inventory change trends) is obtained, and inventory forecast fluctuation data includes but is not limited to forecast information on material shortages and surpluses.

[0062] Based on the first analysis data, the second analysis data, the inventory forecast fluctuation data and the third analysis data, the following Bayesian network algorithm is used to perform root cause analysis: in, is the conditional probability of a defect occurring when the feature vector f is given. is the probability of the feature vector under defect conditions. The prior probability of a defect. The Bayesian network can calculate the conditional dependencies and probability distributions between various factors, and identify the root causes that are most likely to affect production and inventory fluctuations. Specifically, the Bayesian network treats all relevant data as random variables and uses probabilistic inference to determine which factors have the greatest impact on the system. By prioritizing these influencing factors, key issues such as potential bottlenecks, supply chain disruptions, and production delays can be identified. The most likely root cause factors are derived using Bayes' theorem.

[0063] Based on the identified influencing factors, we use the improvement derivation method to propose specific improvement measures. These improvement measures include but are not limited to production line adjustment, inventory optimization, supplier management, procurement strategy adjustment, etc. Through the evaluation and simulation of different improvement measures, we form a set of highly targeted and operational improvement plans (i.e., predicted improvement data).

[0064] For example, for materials with large inventory fluctuations, it may be necessary to alleviate inventory pressure by purchasing in advance, adjusting inventory thresholds, or reaching a more flexible delivery agreement with suppliers; for production bottlenecks, it may be necessary to replan the production line or introduce more advanced production equipment, etc. Ultimately, through improved derivation, "forecast improvement data" was generated, providing enterprises with a clear direction for improvement.

[0065] Step S6: performing material allocation analysis on the anti-disturbance scheduling instructions and the predicted improvement data according to a preset optimization method to obtain real-time warehousing logistics decisions.

[0066] According to the real-time warehouse environment, AGV path conditions and the urgency of the task, the anti-disturbance dispatch instructions are prioritized. Specifically, the system will evaluate the priority of the dispatch instructions based on the following main factors: 1. Consider the current path occupancy of the AGV. If some paths are already congested or are about to be occupied by other tasks, tasks with higher priority will be scheduled to unoccupied paths to avoid path conflicts. 2. The priority of a task is linked to its urgency. For example, some orders may have time requirements, or the production line may have a more urgent need for certain materials, and these tasks will be given a higher priority; 3. Prioritize tasks based on the AGV's operating speed and the estimated completion time of the current task. The tasks that can be completed in a short time will be scheduled with higher priority.

[0067] Each anti-disturbance scheduling instruction will be assigned a priority, thus forming a material scheduling matrix. The material scheduling matrix reflects the priority allocation of tasks in terms of time, path and resources.

[0068] At the same time, through time series analysis, regression analysis or machine learning models, combined with market forecast data, customer order information and seasonal factors, the demand for products is modeled to obtain a demand forecast curve. The demand forecast curve is used to represent the fluctuation of product demand within a certain time range in the future.

[0069] After obtaining the demand forecast curve, the demand forecast curve is compared with the current inventory material data to identify the difference between inventory and demand, and then provide early warning information to avoid excessive inventory or out-of-stock situations. Then, the sorted anti-disturbance scheduling instructions and inventory warning data are combined to generate the final material scheduling instructions. For example, when the inventory warning data shows that certain materials are about to be in short supply, the system prioritizes the allocation of inventory to ensure that urgent needs can be met in a timely manner. By combining material scheduling instructions with path optimization, it is ensured that the path planning of AGVs in the material scheduling process can proceed smoothly, and path conflicts and resource waste can be avoided.

[0070] The AGV path is optimized according to the generated material dispatching instructions and material dispatching matrix to obtain the path optimization instructions. According to the current operation of the AGV, its travel path is dynamically adjusted to avoid path congestion or long waiting time. At the same time, according to the material dispatching matrix and the spatial distribution of tasks, the AGV path is reasonably arranged to ensure that materials can be transported efficiently.

[0071] It should be understood that the task execution list in the digital twin warehouse model will be updated in real time based on the material scheduling instructions and path optimization instructions. The execution status of each task is monitored, so that scheduling adjustments are made according to the material scheduling instructions and path optimization instructions, and real-time warehouse logistics decisions are generated to ensure that each task is completed within a reasonable time. For example, if some tasks are delayed, the system will automatically adjust the relevant paths and scheduling priorities to ensure the efficient operation of the warehouse system.

[0072] This application is applied to the field of warehouse management technology. By preprocessing the target data set to realize standardized spatiotemporal fusion data flow, the virtual space is dynamically mapped based on the standardized data flow to generate a virtual and real synchronized digital twin warehouse model, and multi-target scheduling is performed according to the real-time status data in the digital twin warehouse model to generate an AGV path matrix, and the AGV path matrix is ​​dynamically replanned according to the real-time status data to generate anti-disturbance scheduling instructions. At the same time, the root cause analysis of the product data set is performed to obtain the prediction improvement data, and finally the anti-disturbance scheduling instructions and the prediction improvement data are analyzed for material allocation according to the preset optimization method to obtain the warehouse logistics decision. This application realizes the flexible scheduling of warehouse logistics according to the current situation of warehousing and the prediction analysis of the product cycle, so that the warehousing system can respond to the actual needs of production more accurately, adjust the operation strategy of logistics transportation in time, and effectively respond to various disturbances in the environment, thereby improving the overall warehousing efficiency, reducing costs and improving the intelligent level of warehousing management.

[0073] like Figure 2 The figure shows a functional module diagram of an information management device for intelligent warehousing provided in an embodiment of the present application.

[0074] In some embodiments, the intelligent warehouse information management device 2 may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the intelligent warehouse information management device 2 may be stored in the memory of the server and executed by at least one processor to execute (see Figure 1 (Describe) the functions of the information management method of intelligent warehousing.

[0075] In this embodiment, the information management device 2 of the intelligent warehousing can be divided into multiple functional modules according to the functions it performs. The functional modules may include: a data fusion module 21, a model building module 22, a path matrix module 23, an instruction adjustment module 24, an improved prediction module 25 and a logistics decision module 26. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0076] The data fusion module 21 is used to perform data preprocessing on the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream.

[0077] In an optional implementation, the data fusion module 21 is specifically used for: Collect multi-source heterogeneous data streams in real time to obtain the target data set; Performing time series synchronization processing on the target data set to obtain a time series data set; Performing spatiotemporal fusion processing on the time series data set to obtain a spatiotemporal fusion data stream; Performing data cleaning and interpolation on the spatiotemporal fusion data stream to obtain an advanced spatiotemporal fusion data stream; The advanced spatiotemporal fusion data stream is formatted to obtain the standardized spatiotemporal fusion data stream.

[0078] The model building module 22 is used to perform dynamic mapping of the virtual space according to the standardized space-time fusion data stream to obtain a digital twin warehouse model that is synchronized between the virtual and the real.

[0079] In an optional embodiment, the model building module 22 is specifically used for: Step S21, performing time synchronization and position matching on the real-time sensor data in the standardized spatiotemporal fusion data stream to obtain first synchronized data; Step S22: constructing a virtual space model according to the first synchronization data and a preset warehouse structure data set to obtain a primary space model; Step S23, performing layered processing on the primary space model to obtain an intermediate space model; Step S24: Mapping the real-time status data and environmental data in the standardized spatiotemporal fusion data stream to the intermediate space model to obtain a digital twin warehouse model at the current moment; Step S24 is repeatedly executed according to the standardized spatiotemporal fusion data stream updated in real time to update the digital twin warehouse model in real time.

[0080] The path matrix module 23 is used to perform multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix.

[0081] In an optional implementation, the path matrix module 23 is specifically used for: Prioritizing each warehousing task point in the digital twin warehousing model to obtain a task execution list; Generating a plurality of potential path sets according to a preset constraint optimization model and the task execution list; Obtain the optimal path corresponding to each storage task point according to the preset multi-objective optimization trade-off algorithm to obtain the optimal path combination; Tasks are assigned to each AGV according to the optimal path combination and the task execution list to obtain the AGV path matrix.

[0082] The instruction adjustment module 24 is used to dynamically replan the AGV path matrix according to the real-time status data of the AGV equipment to obtain anti-disturbance scheduling instructions.

[0083] In an optional implementation, the instruction adjustment module 24 is specifically used to: Acquire real-time status data of each AGV device according to the AGV path matrix to perform abnormality detection according to the real-time status data; Dynamically adjust the AGV path matrix according to a preset model control prediction algorithm and detected abnormal information; The anti-disturbance scheduling instruction is generated according to the adjusted AGV path matrix and the task execution list.

[0084] The improvement prediction module 25 is used to obtain a product data set of a target system and perform root cause analysis on the product data set to obtain prediction improvement data.

[0085] In an optional implementation, the improved prediction module 25 is specifically used for: Acquire a product data set of a target system and identify production requirements according to the product data set to obtain first analysis data; Performing capacity analysis according to the product data set to obtain second analysis data; Performing matching analysis on the inventory material data in the target data set according to the first analysis data to obtain inventory forecast fluctuation data; Perform product cycle analysis according to the product data set to obtain third analysis data; Performing root cause analysis on the first analysis data, the second analysis data, the inventory forecast fluctuation data, and the third analysis data according to a preset Bayesian network algorithm to identify potential influencing factors; An improved derivation is performed on each influencing factor to obtain the prediction improvement data.

[0086] The logistics decision module 26 is used to perform material allocation analysis on the anti-disturbance scheduling instructions and the predicted improvement data according to a preset optimization method to obtain real-time warehousing logistics decisions.

[0087] In an optional embodiment, the logistics decision module 26 is specifically used for: Prioritizing the anti-disturbance scheduling instructions to obtain a material scheduling matrix; Performing demand fluctuation analysis on the forecast improvement data to obtain a demand forecast curve; Comparing the demand forecast curve with the inventory material data to obtain inventory warning data; Generate a material dispatch instruction according to the anti-disturbance dispatch instruction and the inventory warning data; Generate a path optimization instruction according to the material dispatch instruction and the material dispatch matrix; The task execution list in the digital twin warehousing model is feedback-scheduled according to the material scheduling instructions and the path optimization instructions to obtain the real-time warehousing logistics decision.

[0088] It should be understood that the various variations and specific embodiments of the methods provided in the above embodiments are also applicable to the information management device for intelligent warehousing in the present embodiment. Through the above detailed description of the information management method for intelligent warehousing, those skilled in the art can clearly know the implementation method of the information management device for intelligent warehousing in the present embodiment. For the sake of brevity of the specification, it will not be described in detail here.

[0089] like Figure 3 , which is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0090] In a preferred embodiment of the present invention, the electronic device 3 may include, but is not limited to: a memory 31 , at least one processor 32 and at least one communication bus 33 .

[0091] Those skilled in the art should understand that Figure 3 The structure of the electronic device 3 shown does not constitute a limitation of the embodiment of the present invention, and the electronic device 3 may also include more or less other hardware or software than shown in the figure, or a different arrangement of components.

[0092] In some embodiments, the electronic device 3 is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits, programmable gate arrays, digital processors and embedded devices.

[0093] It should be noted that the electronic device 3 is only an example, and other existing or future electronic products that are suitable for the present application should also be included in the protection scope of the present application and included here by reference.

[0094] In some embodiments, the memory 31 stores a computer program, and when the computer program is executed by the at least one processor 32, all or part of the steps in the information management method of intelligent warehousing as described above are implemented. The memory 31 includes a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data. Further, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, and the like.

[0095] In some embodiments, the at least one processor 32 is the control core (ControlUnit) of the electronic device 3, and uses various interfaces and lines to connect various components of the entire electronic device 3, and executes various functions and processes data of the electronic device 3 by running or executing programs or modules stored in the memory 31, and calling data stored in the memory 31. For example, when the at least one processor 32 executes the computer program stored in the memory 31, it implements all or part of the steps of the information management method of intelligent warehousing described in the embodiment of the present application; or implements all or part of the functions of the information management device of intelligent warehousing. The at least one processor 32 can be composed of an integrated circuit, for example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits with the same function or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors, and various control chips.

[0096] In some embodiments, the at least one communication bus 33 is configured to realize the connection and communication between the memory 31 and the at least one processor 32. Although not shown, the electronic device 3 may also include a power supply (such as a battery) for powering each component. Preferably, the power supply may be logically connected to the at least one processor 32 through a power management device, so as to realize the functions of managing charging, discharging, and power consumption management through the power management device. The power supply may also include any components such as one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device 3 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0097] The above-mentioned integrated unit implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling an electronic device (which can be a personal computer, electronic device, or network device, etc.) or a processor to execute part of the method described in each embodiment of the present application.

[0098] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0099] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0100] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. An information management method for intelligent warehousing, characterized in that: The method comprises: Perform data preprocessing on the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream; Performing dynamic mapping of virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model; Performing multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix; Dynamically replanning the AGV path matrix according to the real-time status data of the AGV device to obtain anti-disturbance scheduling instructions; Acquire a product data set of a target system and perform root cause analysis on the product data set to obtain prediction improvement data; The anti-disturbance scheduling instructions and the predicted improvement data are analyzed for material allocation according to a preset optimization method to obtain real-time warehousing logistics decisions.

2. The information management method for intelligent warehousing according to claim 1 is characterized in that: The data preprocessing of the target data set collected in real time according to the preset data processing method to obtain a standardized spatiotemporal fusion data stream includes: Collect multi-source heterogeneous data streams in real time to obtain the target data set; Performing time series synchronization processing on the target data set to obtain a time series data set; Performing spatiotemporal fusion processing on the time series data set to obtain a spatiotemporal fusion data stream; Performing data cleaning and interpolation on the spatiotemporal fusion data stream to obtain an advanced spatiotemporal fusion data stream; The advanced spatiotemporal fusion data stream is formatted to obtain the standardized spatiotemporal fusion data stream.

3. The information management method for intelligent warehousing according to claim 1 is characterized in that: The method of dynamically mapping the virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model includes: Step S21, performing time synchronization and position matching on the real-time sensor data in the standardized spatiotemporal fusion data stream to obtain first synchronized data; Step S22: constructing a virtual space model according to the first synchronization data and a preset warehouse structure data set to obtain a primary space model; Step S23, performing layered processing on the primary space model to obtain an intermediate space model; Step S24: Mapping the real-time status data and environmental data in the standardized spatiotemporal fusion data stream to the intermediate space model to obtain a digital twin warehouse model at the current moment; Step S24 is repeatedly executed according to the standardized spatiotemporal fusion data stream updated in real time to update the digital twin warehouse model in real time.

4. The information management method for intelligent warehousing according to claim 1 is characterized in that: The multi-objective scheduling process of the digital twin warehouse model to obtain the AGV path matrix includes: Prioritizing each warehousing task point in the digital twin warehousing model to obtain a task execution list; Generating a plurality of potential path sets according to a preset constraint optimization model and the task execution list; Obtain the optimal path corresponding to each storage task point according to the preset multi-objective optimization trade-off algorithm to obtain the optimal path combination; Tasks are assigned to each AGV according to the optimal path combination and the task execution list to obtain the AGV path matrix.

5. The information management method for intelligent warehousing according to claim 4 is characterized in that: The dynamically replanning the AGV path matrix according to the real-time status data of the AGV device to obtain the anti-disturbance scheduling instruction includes: Acquire real-time status data of each AGV device according to the AGV path matrix to perform abnormality detection according to the real-time status data; Dynamically adjust the AGV path matrix according to a preset model control prediction algorithm and detected abnormal information; The anti-disturbance scheduling instruction is generated according to the adjusted AGV path matrix and the task execution list.

6. The information management method for intelligent warehousing according to claim 1 is characterized in that: The acquiring of a product data set of a target system and performing root cause analysis on the product data set to obtain prediction improvement data includes: Acquire a product data set of a target system and identify production requirements according to the product data set to obtain first analysis data; Performing capacity analysis according to the product data set to obtain second analysis data; Performing matching analysis on the inventory material data in the target data set according to the first analysis data to obtain inventory forecast fluctuation data; Perform product cycle analysis according to the product data set to obtain third analysis data; Performing root cause analysis on the first analysis data, the second analysis data, the inventory forecast fluctuation data, and the third analysis data according to a preset Bayesian network algorithm to identify potential influencing factors; An improved derivation is performed on each influencing factor to obtain the prediction improvement data.

7. The information management method for intelligent warehousing according to claim 6 is characterized in that: The material allocation analysis of the anti-disturbance scheduling instructions and the prediction improvement data according to the preset optimization method to obtain real-time warehousing logistics decisions includes: Prioritizing the anti-disturbance scheduling instructions to obtain a material scheduling matrix; Performing demand fluctuation analysis on the forecast improvement data to obtain a demand forecast curve; Comparing the demand forecast curve with the inventory material data to obtain inventory warning data; Generate a material dispatch instruction according to the anti-disturbance dispatch instruction and the inventory warning data; Generate a path optimization instruction according to the material dispatch instruction and the material dispatch matrix; The task execution list in the digital twin warehousing model is feedback-scheduled according to the material scheduling instructions and the path optimization instructions to obtain the real-time warehousing logistics decision.

8. An information management device for intelligent warehousing, characterized in that: The device comprises: A data fusion module is used to perform data preprocessing on the target data set collected in real time according to a preset data processing method to obtain a standardized spatiotemporal fusion data stream; A model building module, used for dynamically mapping virtual space according to the standardized spatiotemporal fusion data stream to obtain a virtual-real synchronized digital twin warehouse model; A path matrix module is used to perform multi-objective scheduling processing on the digital twin warehouse model to obtain an AGV path matrix; An instruction adjustment module is used to dynamically replan the AGV path matrix according to the real-time status data of the AGV equipment to obtain anti-disturbance scheduling instructions; An improvement prediction module, used for acquiring a product data set of a target system and performing root cause analysis on the product data set to obtain prediction improvement data; The logistics decision module is used to perform material allocation analysis on the anti-disturbance scheduling instructions and the predicted improvement data according to a preset optimization method to obtain real-time warehousing logistics decisions.

9. An electronic device, characterized in that: The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the information management method for intelligent warehousing according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the information management method of intelligent warehousing according to any one of claims 1 to 7 are implemented.

Images