A big data-based intelligent factory warehouse management method and system

Abstract

The application discloses a big data-based intelligent factory warehouse management method and system. The method comprises the following steps: quantifying the demand change trend of goods by determining the dynamic position distribution of the goods and calculating the access frequency fluctuation of the goods; dividing the goods into two categories of high-frequency goods and low-frequency goods according to the quantified value of the demand change trend of the goods, so as to determine the area range of the optimized storage of the high-frequency goods; determining the layout scheme of the storage position and the optimal goods taking path sequence without conflict through the area range of the optimized storage of the high-frequency goods, so as to obtain an optimized resource allocation scheme. The method can improve the efficiency and flexibility of the warehouse system.

Description

Technical Field

[0001] This application relates to the field of smart factory warehouse management technology, and in particular to a smart factory warehouse management method and system based on big data. Background Technology

[0002] Smart factory warehouse management, as a key area for improving supply chain efficiency and enterprise competitiveness, is receiving increasing attention. Its core lies in using data-driven methods to achieve visualization and optimized scheduling of the entire product lifecycle, thereby significantly reducing operating costs and improving response speed. However, traditional warehouse management methods are gradually revealing their limitations when facing increasingly complex production demands. These methods often rely on static storage planning and human experience, making it difficult to adapt to dynamically changing market demands and failing to fully utilize real-time data to optimize product layout and retrieval efficiency. This results in a dual deficiency in efficiency and flexibility of the warehouse system, particularly in balancing the rapid response of high-frequency goods with the rational storage of low-frequency goods. Summary of the Invention

[0003] This application provides a smart factory warehouse management method and system based on big data to improve the efficiency and flexibility of the warehouse system.

[0004] Firstly, in order to solve the aforementioned technical problems, this application provides a smart factory warehouse management method based on big data, including: Each item is tagged with an RFID code, and the items are bound to a data acquisition device equipped with sensors to obtain initial location data of the items at the warehouse entrance and record it in the database, thereby obtaining a preliminary location mapping. Based on the preliminary location mapping, the movement trajectory of the goods from the warehouse to the storage point is analyzed by combining the positioning algorithm with multi-point sensor data to calculate the path node density, thereby determining the current dynamic location distribution of the goods. Based on the dynamic location distribution, historical pickup data within the classification time window are fused using time series processing technology to calculate the pickup frequency fluctuation, thereby obtaining a quantitative value of the demand change trend of the goods. Determine whether the quantitative value of the demand change trend exceeds the usage frequency threshold. If the quantitative value of the demand change trend exceeds the usage frequency threshold, use a clustering algorithm combined with usage frequency fluctuation and cargo volume parameters to divide the cargo into high-frequency and low-frequency categories, thereby determining the optimal storage area for high-frequency cargo. Based on the optimized storage area of ​​the high-frequency goods, the code association data is obtained through the goods identification and incorporated into dynamic storage to obtain the storage location adjustment requirements; Based on the adjustment requirements of the storage location, the storage location is replanned by dividing it into regional granularities using a simulated annealing algorithm and combining it with cargo volume parameters to obtain the adjusted layout scheme. Based on the adjusted layout scheme, the A* algorithm is used to generate pickup paths by combining path node density and real-time speed data collection. At the same time, the path intersection probability and conflict range are detected to obtain the optimal pickup path sequence without conflict. The optimal picking path sequence guides the handling equipment to perform picking operations, and real-time speed data and the location coordinates of dynamic obstacles are obtained from the operation log to determine the accuracy of path intersection probability detection and update the conflict detection range. The path length changes are analyzed based on the updated conflict detection range, and the storage layout is adjusted by using frequency fluctuations and dynamic obstacle coordinates to obtain an optimized resource allocation scheme.

[0005] Secondly, this application provides a storage medium storing computer program data, which, when executed, enables the implementation of the big data-based smart factory warehouse management method described above.

[0006] Thirdly, this application provides a smart factory and warehouse management system based on big data, including: The preliminary location mapping module is used to assign an identification code to each item of goods using RFID technology, and to obtain the initial location data of the goods at the warehouse entrance by binding the goods to a data acquisition device equipped with sensors and recording it in the database, thereby obtaining the preliminary location mapping. The dynamic location distribution determination module is used to analyze the movement trajectory of the goods from the warehouse to the storage point by combining the positioning algorithm with multi-point sensor data based on the preliminary location mapping, so as to calculate the path node density and thus determine the current dynamic location distribution of the goods. The demand change trend quantification value acquisition module is used to calculate the frequency fluctuation of the pickup based on the dynamic location distribution and by using time series processing technology to fuse historical pickup data within the classification time window, thereby obtaining the quantification value of the demand change trend of the goods. The module for determining the optimal storage area range for high-frequency goods is used to determine whether the quantitative value of the demand change trend exceeds the retrieval frequency threshold. When the quantitative value of the demand change trend exceeds the retrieval frequency threshold, the module uses a clustering algorithm combined with retrieval frequency fluctuations and goods volume parameters to divide the goods into two categories: high-frequency and low-frequency, thereby determining the optimal storage area range for high-frequency goods. The storage location adjustment requirement acquisition module is used to obtain the code association data through the cargo identifier and integrate it into dynamic storage based on the optimized storage area range of the high-frequency goods, so as to obtain the storage location adjustment requirements. The adjusted layout scheme acquisition module is used to re-plan the storage location according to the adjustment requirements of the storage location by dividing the granularity by region using the simulated annealing algorithm and combining it with the cargo volume parameters, so as to obtain the adjusted layout scheme. The conflict-free optimal pickup path sequence acquisition module is used to generate pickup paths based on the adjusted layout scheme by combining the A* algorithm with path node density and real-time speed data acquisition, and at the same time detect the path intersection probability and conflict range to obtain the conflict-free optimal pickup path sequence. The conflict detection range update module is used to guide the handling equipment to perform the picking operation according to the optimal picking path sequence, and to obtain real-time speed data and the position coordinates of dynamic obstacles from the operation log in order to determine the accuracy of the path intersection probability detection and update the conflict detection range. The optimized resource allocation scheme acquisition module is used to analyze path length changes based on the updated conflict detection range, and adjust the storage layout by using frequency fluctuations and dynamic obstacle coordinates to obtain the optimized resource allocation scheme.

[0007] Compared with existing technologies, this application has at least the following beneficial effects: by determining the dynamic location distribution of goods and calculating the fluctuation of goods retrieval frequency, the demand change trend of goods can be quantified; further, by using the quantified value of the demand change trend of goods, goods are divided into high-frequency and low-frequency categories to determine the optimal storage area for high-frequency goods; further, by using the optimal storage area for high-frequency goods, the layout scheme of storage locations and the optimal retrieval path sequence without conflict can be determined, thereby obtaining an optimized resource allocation scheme, and thus improving the efficiency and flexibility of the warehousing system. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating the smart factory warehouse management method based on big data provided in the first embodiment of this application; Figure 2 This is a schematic diagram of the structure of the smart factory warehouse management system based on big data provided in the second embodiment of this application. Detailed Implementation

[0009] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0010] Reference Figure 1 The first embodiment of this application provides a smart factory warehouse management method based on big data, including the following steps: S101: Each item is tagged with an RFID code and its initial location data is obtained at the warehouse entrance and recorded in the database by binding the item to a sensor-equipped data acquisition device, thereby obtaining a preliminary location mapping. S102: Based on the initial location mapping, the movement trajectory of goods from the warehouse to the storage point is analyzed by combining the positioning algorithm with multi-point sensor data to calculate the path node density, thereby determining the current dynamic location distribution of the goods; S103: Based on the dynamic location distribution, use time series processing technology to fuse historical pickup data within the classification time window to calculate the pickup frequency fluctuation, thereby obtaining a quantitative value of the demand trend of goods. S104: Determine whether the quantitative value of the demand change trend exceeds the usage frequency threshold. When the quantitative value of the demand change trend exceeds the usage frequency threshold, use a clustering algorithm combined with usage frequency fluctuations and cargo volume parameters to divide the cargo into high-frequency and low-frequency categories, thereby determining the optimal storage area for high-frequency cargo. S105: Based on the optimized storage area of ​​high-frequency goods, obtain the code association data through the goods identification and integrate it into dynamic storage to obtain the storage location adjustment requirements; S106: Based on the need to adjust the storage location, the storage location is replanned by dividing the granularity of the region using the simulated annealing algorithm and combining it with the cargo volume parameters to obtain the adjusted layout scheme. S107: Based on the adjusted layout scheme, the pickup path is generated by combining the A* algorithm with path node density and real-time speed data. At the same time, the path intersection probability and conflict range are detected to obtain the optimal pickup path sequence without conflict. S108: Guide the handling equipment to perform picking operations based on the optimal picking path sequence, and obtain real-time speed data and the location coordinates of dynamic obstacles from the operation log to determine the accuracy of path intersection probability detection and update the conflict detection range; S109: Analyze path length changes based on the updated conflict detection range, and adjust the storage layout by using frequency fluctuations and dynamic obstacle coordinates to obtain an optimized resource allocation scheme.

[0011] In step S101, each item is assigned an identification code using RFID technology, and the item is bound to a data acquisition device equipped with sensors to obtain the initial location data of the item at the warehouse entrance and record it in the database, thereby obtaining a preliminary location mapping.

[0012] Cargo identification tags are generated using sensors and RFID technology and bound to a data collection device. Initial location data is acquired from the warehouse entrance and entered into a database to obtain a preliminary location mapping. The binding relationship between location information and the data collection device is used to determine if the cargo identification tag matches the initial location. If they match, the location mapping is updated using the database record. The updated location mapping is then used to detect location information deviations through precision constraints. If the deviation exceeds a preset threshold, the K-means algorithm is used to cluster the data from the data collection device to identify abnormal location points. Based on these abnormal location points, the corresponding cargo identification tags are extracted from the database records. RFID technology is used to verify the matching status between the cargo identification tag and the location information to obtain corrected location data. Using the corrected location data, real-time location information is acquired using sensor technology to determine if it matches the database record. If they do not match, the database record is updated and a new location mapping is generated.

[0013] Specifically, generating cargo identification tags and binding them to data collection devices using sensors and RFID technology is the foundation for cargo tracking in modern warehouse management.

[0014] For example, in a large logistics warehouse, each item is uniquely identified upon arrival using an RFID tag, such as "item A-001," and linked to a data collection device equipped with sensors. This device can be a reader / writer fixed to a shelf or a mobile handheld terminal. Initial location data is obtained from the warehouse entrance; for example, item A-001 is recorded as "shelf 1 - zone A - position 01" and entered into the database, forming a preliminary location mapping. This step ensures an initial association between the item and its location, laying the foundation for subsequent tracking.

[0015] In one possible implementation, the binding relationship between location information and the data acquisition device is used to determine whether the cargo identification is consistent with the initial location.

[0016] Specifically, assuming the data acquisition device for item A-001 continuously reports the location as "Shelf 1 - Zone A - Position 01," consistent with the database record, the location mapping is updated through the database to maintain data real-time performance. This consistency verification effectively reduces location errors caused by human error, improving the accuracy of inventory management. After obtaining the updated location mapping, location information deviations are detected through precision constraints.

[0017] For example, a preset threshold can be set, such as the position deviation not exceeding 50 centimeters. If the location information of goods A-001 is displayed as "Shelf 1 - Zone A - Position 02", the deviation exceeds the threshold, indicating that there may be an anomaly.

[0018] Preferably, the K-means algorithm is used to perform cluster analysis on the data collected by the acquisition devices. For example, the location data returned by multiple acquisition devices in the warehouse are divided into three categories, one of which is concentrated near "Shelf 1 - Area A - Position 02", indicating that this is an abnormal location. This clustering method quickly locates the problem area by grouping data, improving the efficiency of anomaly detection. Based on the abnormal location points, the corresponding cargo identifiers are extracted from the database, such as cargo A-001, and the matching status with the location information is verified using RFID technology.

[0019] For example, if an RFID scan shows that item A-001 is actually located at "Shelf 1 - Zone A - Position 02" instead of the initially recorded "Position 01", the location data is corrected to the latest scan result. This verification method utilizes the high precision of RFID to ensure the reliability of location data and avoid confusion caused by location drift. Real-time location information is then obtained using the corrected location data combined with sensor technology.

[0020] For example, if the sensor updates the location of item A-001 every 5 minutes and displays "Shelf 1 - Zone A - Position 02", which matches the corrected data, no further action is needed. However, if it displays "Shelf 1 - Zone B - Position 01", it indicates that the item has been moved, requiring an update to the database and the generation of a new location mapping.

[0021] Understandably, this real-time update mechanism can reflect the actual dynamics of goods in a timely manner, making it easier for managers to respond quickly to inventory changes.

[0022] It should be noted that the above process achieves closed-loop management from cargo identification generation to location mapping update through the collaborative work of sensors and RFID.

[0023] In one embodiment, assuming the warehouse processes 1,000 items per day, deviation detection and anomaly location can reduce the location error rate from 5% to 1%, significantly improving warehousing efficiency.

[0024] For example, managers can quickly identify abnormal goods and adjust the layout through system reports, reducing the time spent searching for goods. This technology not only optimizes operational processes but also provides data support for large-scale warehousing, enhancing overall management capabilities.

[0025] In step S102, based on the preliminary location mapping, the movement trajectory of the goods from the warehouse to the storage point is analyzed using a positioning algorithm combined with multi-point sensor data to calculate the path node density, thereby determining the current dynamic location distribution of the goods. Following this, the following steps are also included: S1021: Verify the matching status between the location distribution of goods and storage points through multi-point analysis; Specifically, by fusing sensor data through positioning algorithms to analyze the movement trajectory of goods from the warehouse to the storage point, it can be understood as a method based on multi-source data integration.

[0026] For example, in a large warehouse, after goods enter through the entrance, sensors continuously record their location information. For instance, infrared sensors capture the coordinates of the points the goods pass through, and combined with the identification data read by RFID, a complete movement path is generated through a positioning algorithm.

[0027] For example, the trajectory of cargo B-002 from the entrance to "Shelf 2 - Zone B - Position 03" may include passing through four nodes: "Entrance - Aisle 1 - Transfer Area - Shelf 2", with a path length of approximately 20 meters. Node density data is extracted from the distribution of path nodes, and statistical tools are used to calculate density variation characteristics, aiming to reveal concentrated areas of cargo movement.

[0028] Specifically, suppose 500 goods pass through Channel 1 in a day, which has a high node density, while only 100 goods pass through the transit area, which has a low density. Statistical tools, such as mean analysis, can reveal that the density of Channel 1 changes steadily, while the density of the transit area fluctuates more significantly. This helps in inferring the current dynamic location of the goods.

[0029] For example, if cargo B-002 remains in the transit area, it may indicate that its movement is not yet complete. After obtaining the dynamic location, multi-point analysis is used to verify the matching status between the location distribution and the storage points to determine if any deviations have occurred.

[0030] S1022: Determine whether there is a deviation between the location distribution and the storage points; if a deviation exists, cluster the sensor data using the K-means algorithm to obtain a set of abnormal nodes; In one possible implementation, the system compares the real-time location of cargo B-002, "transfer area," with the target storage point, "shelf 2 - section B - position 03." If a discrepancy is found, it indicates that the cargo has not arrived as expected.

[0031] Preferably, multi-point analysis can utilize data from multiple sensors, such as readers at the entrance and on shelves, to comprehensively determine the location matching degree. If discrepancies exist, the sensor data is clustered using the K-means algorithm to obtain a set of abnormal nodes.

[0032] For example, suppose the data from 100 sensors in a warehouse is divided into three categories. One category shows that goods have deviated from "Shelf 1-C" by more than 5 meters from their target storage point. This type of data constitutes an abnormal node set. This clustering method can quickly pinpoint the problem area.

[0033] S1023: Based on the set of abnormal nodes, extract the corresponding movement trajectory segments from the trajectory analysis to determine the time range of the abnormality; Based on the set of abnormal nodes, the corresponding movement trajectory segments are extracted from the trajectory analysis to determine the time range of the anomaly.

[0034] For example, if the trajectory of cargo B-002 shows that it entered the transfer area from channel 1 at 10:00 am, but its location is abnormally located in "shelf 1-C area" after 10:30 am, then the abnormal time range is from 10:00 am to 10:30 am.

[0035] It should be noted that this time range extraction can help trace the cause of problems, such as human error in handling.

[0036] S1024: By using sensor data within the time range of the anomaly occurrence, combined with a positioning algorithm, the dynamic position is corrected to obtain the adjusted position distribution of the goods; By combining sensor data within a time range with positioning algorithms to correct dynamic positions, the adjusted position distribution is obtained.

[0037] In one embodiment, the system analyzes sensor records from 10:00 to 10:30 and finds that item B-002 has been mistakenly placed in "Shelf 1 - Zone C". Subsequently, an algorithm adjusts its position to the actual scan point "Shelf 2 - Zone B - Position 03". This correction ensures data accuracy.

[0038] S1025: Based on the adjusted location distribution, update the current dynamic location distribution of goods by comparing the stored point data.

[0039] Obtain the adjusted location distribution and update the current dynamic location record of the goods by comparing the stored point data.

[0040] For example, the original record in the database for goods B-002, which was "transfer area", has now been updated to "shelf 2 - section B - position 03".

[0041] Understandably, this real-time update mechanism can reflect the status of goods in a timely manner, which facilitates subsequent management.

[0042] In step S103, based on the dynamic location distribution, historical pickup data within the classification time window is fused using time series processing technology to calculate the pickup frequency fluctuation, thereby obtaining a quantitative value of the demand change trend of goods, including the following steps: S1031: Extract the storage location information of goods based on the dynamic location distribution to establish a location index table; Specifically, the process of extracting cargo storage location information and establishing a location index table based on dynamic location distribution data can be understood as a systematic data processing method.

[0043] For example, in a large warehouse, the storage location information of goods may include specific coordinates such as "shelf 2 - zone B - position 03". By analyzing dynamic location distribution data, the system will summarize the current location of all goods to form an index table containing location numbers and goods identifiers.

[0044] For example, item B-002 corresponds to "shelf 2 - section B - position 03", and item C-005 corresponds to "shelf 3 - section A - position 01". This index table facilitates quick location of goods.

[0045] S1032: Using a sliding time window method, retrieve pickup records within a specified time range from the historical database, and calculate the pickup frequency value corresponding to each location in the location index table through a frequency analysis algorithm; Using a sliding time window method to retrieve pickup records within a specified time range from a historical database is a dynamic analysis technique.

[0046] Specifically, assuming a 24-hour time window, the system retrieves all pickup records from the database for the past day. For example, goods B-002 were picked up at 10:00 AM, and goods C-005 were picked up at 2:00 PM. By using a sliding window, the time range can be progressively moved forward, continuously updating the data. This method captures the temporal characteristics of pickup behavior. A frequency analysis algorithm calculates the pickup frequency value corresponding to each location index, aiming to quantify the location's activity level.

[0047] S1033: Determine whether the frequency value exceeds the preset threshold. If the frequency value exceeds the preset threshold, mark the location as a high-frequency retrieval area. In one possible implementation, the system counts that "Shelf 2 - Zone B - Position 03" was used 5 times within 24 hours, and "Shelf 3 - Zone A - Position 01" was used 2 times, calculating the frequency values ​​for each. If the preset threshold is 4 times, then "Shelf 2 - Zone B - Position 03" exceeds the threshold and is marked as a high-frequency usage area.

[0048] It should be noted that this marking can highlight areas that are frequently used, making it easier to optimize management.

[0049] S1034: Based on the distribution of high-frequency access areas, cluster analysis is used to divide the storage areas of goods; Based on the distribution of high-frequency access areas, cluster analysis is used to divide the cargo storage areas, which enables spatial partitioning management.

[0050] Preferably, assuming there are 10 high-frequency retrieval locations in the warehouse, cluster analysis may divide them into 3 regions: region A near the entrance, region B in the middle aisle, and region C deep in the shelving.

[0051] For example, Zone A includes "Shelf 2 - Section B" because it is frequently accessed and located in a concentrated area. This division helps to rationally plan the storage layout.

[0052] S1035: Using time series processing technology, volatility analysis is performed on the access frequency of each storage area, and based on the results of the volatility analysis, a regression model is used to quantify the demand change trend in order to obtain a quantitative value of the demand change trend of goods.

[0053] By using time series processing techniques to analyze the volatility of usage frequency in different regions, the regularity of usage patterns can be revealed.

[0054] In one embodiment, the system analyzes the usage frequency of region A over the past 7 days, finding larger fluctuations on Mondays and Fridays (8 and 10 times respectively), while Wednesdays saw only 3 usages. This volatility analysis reflects the temporal characteristics of demand changes. Based on the volatility analysis results, a regression model is used to quantify the demand change trend and obtain trend predictions, representing a forward-looking analytical approach.

[0055] For example, based on 7 days of data from region A, a regression model predicts that the frequency of retrieval may rise to 12 times in the next 3 days. This prediction can provide a basis for inventory adjustments.

[0056] Understandably, accurate trend forecasting helps in preparing resources in advance and improving efficiency.

[0057] In step S104, it is determined whether the quantified value of the demand change trend exceeds the usage frequency threshold. If the quantified value of the demand change trend exceeds the usage frequency threshold, a clustering algorithm is used to combine usage frequency fluctuations and cargo volume parameters to classify the cargo into high-frequency and low-frequency categories, thereby determining the optimal storage area for high-frequency cargo. This includes the following steps: S1041: Obtain the quantitative value of demand change trend and usage frequency data through historical usage records, and determine whether the quantitative value exceeds the preset usage frequency threshold to obtain an over-threshold marker. Specifically, obtaining quantitative values ​​of demand changes and usage frequency data through historical usage records can be understood as a data-driven dynamic evaluation method.

[0058] Specifically, the system extracts the pickup records from the warehouse database for the past 30 days. For example, goods A-001 are picked up 3 times a day, and goods B-002 are picked up 8 times a day, forming quantifiable values. The preset pickup frequency threshold is 5 times. Goods B-002 exceeds the threshold and is marked as exceeding the threshold, while A-001 does not reach the threshold and is not marked. This judgment logic is clear and can quickly filter out active goods.

[0059] S1042: Extract frequency fluctuation data based on the threshold markers, integrate the volume parameters of the goods, and use a clustering algorithm to divide high-frequency goods into low-frequency goods to determine the classification results; For example, frequency fluctuation data is extracted based on the threshold mark and the cargo volume parameter is fused, and a clustering algorithm is used to divide high-frequency and low-frequency cargo.

[0060] In one possible implementation, cargo B-002 is retrieved 6 to 10 times per day, with a volume of 0.5 cubic meters; cargo C-003 is retrieved 1 to 3 times per day, with a volume of 1.2 cubic meters. The clustering algorithm combines frequency and volume dimensions, classifying B-002 as a high-frequency cargo and C-003 as a low-frequency cargo. This classification considers not only retrieval activity but also differences in space occupancy.

[0061] S1043: For high-frequency goods in the classification results, calculate the fluctuation amplitude through frequency fluctuation analysis to obtain the priority ranking of high-frequency goods; For high-frequency cargo, frequency fluctuation analysis is performed to calculate the fluctuation amplitude and determine the priority ranking.

[0062] For example, B-002 was used 8, 10, 6, 9, 7, 8, 10 times in the past 7 days, showing significant fluctuations and indicating unstable but generally high demand. Another high-frequency commodity, D-004, was used 7, 8, 7, 7, 8, 7, 8 times, showing small and stable fluctuations. In the priority ranking, B-002 is ranked first due to its high frequency and significant fluctuations, followed by D-004. This ranking highlights the urgency of demand and facilitates priority processing.

[0063] S1044: By combining priority sorting with cargo volume parameters, calculate the granularity of area division to determine the optimal storage area range for high-frequency cargo.

[0064] Preferably, the granularity of area division is calculated by combining priority ranking and cargo volume data to determine the optimal storage range.

[0065] In one embodiment, B-002, though small in size but with high priority, is suitable for area X near the entrance, occupying 0.5 square meters; D-004, with a volume of 0.8 cubic meters and slightly lower priority, is allocated to area Y near the passageway, occupying 1 square meter. The granularity of area division is adjusted based on the characteristics of the goods; for example, the total area occupied in entrance area X is limited to 2 square meters to ensure that frequently used goods are stored centrally. This approach improves retrieval efficiency.

[0066] It should be noted that clustering based on volume parameters can avoid spatial imbalance caused by relying solely on frequency.

[0067] For example, if only frequency is considered, the large C-003 might take up space for high-frequency, small-volume goods, affecting operational smoothness. Adding the volume dimension makes the classification more reasonable.

[0068] Understandably, fluctuation analysis provides a basis for prioritization, making resource allocation more accurate.

[0069] In one possible implementation, after high-frequency goods are concentrated in area X, the pickup time is reduced from 5 minutes to 3 minutes, resulting in a significant improvement in efficiency.

[0070] Specifically, the determination of the optimal storage area can also be adjusted based on the warehouse layout.

[0071] For example, area X is close to the forklift aisle, making it suitable for rapid material handling; area Y is adjacent to the packaging area, facilitating subsequent process connections. This zoning logic can make full use of spatial characteristics and enhance operational continuity.

[0072] For example, if a warehouse processes 500 items per day, the overall process time is reduced by 10% after the high-frequency items are centrally optimized, which is a significant effect.

[0073] In step S105, based on the optimized storage area of ​​high-frequency goods, the code association data is obtained through the goods identification and integrated into dynamic storage to obtain the storage location adjustment requirements; By obtaining code-related data from cargo identification and integrating it with dynamic storage information, an initial layout plan is determined. The code-related data is then used to analyze dynamic storage change trends, revealing the need for adjustments to storage locations.

[0074] Specifically, the code-related data is obtained through cargo identification, and dynamic storage information is integrated to determine the initial layout scheme.

[0075] It is understandable that cargo identifiers such as A-001 and B-002 usually correspond to unique codes, and the associated data of the codes includes information such as cargo type and warehousing time.

[0076] For example, code A-001 indicates it is a small part, having been received 3 times in the past 30 days, while B-002 is a fast-moving consumer good, having been received 10 times. After integrating dynamic storage information, the initial layout plan temporarily places A-001 further back in the storage and B-002 closer to the entrance / exit, reflecting their difference in activity levels. Analyzing the dynamic storage change trends using code-related data reveals the need for adjusting storage locations.

[0077] In step S106, based on the storage location adjustment requirements, the storage locations are replanned by dividing the space into regions using a simulated annealing algorithm and combining it with cargo volume parameters to obtain an adjusted layout scheme, including the following steps: S1061: Extract the granularity of region division based on the storage location adjustment requirements and layout information; Specifically, the system tracked that the frequency of B-002's inbound shipments increased from 5 times per week to 10 times per week, indicating rising demand and a need to relocate to a more convenient location. Meanwhile, A-001's changes remained stable, and its location did not require immediate adjustment. This trend analysis relies on historical data, highlighting the dynamic characteristics of the goods. Based on the adjustment needs and layout information, regional division data is extracted to determine whether the granularity of the division meets storage requirements.

[0078] For example, the warehouse is divided into an entrance area X and a back-end area Y, with an initial granularity of 5 square meters per area. B-002 requires 2 square meters to enter the entrance area X, but if the number of high-frequency goods increases, 5 square meters may not be enough.

[0079] S1062: Determine whether the region partitioning granularity meets the storage requirements. If the region partitioning granularity does not meet the storage requirements, adjust the region partitioning granularity through simulated annealing algorithm to obtain the optimized partitioning result. It should be noted that if the conditions are not met, the partitioning granularity is adjusted using a simulated annealing algorithm.

[0080] In one possible implementation, the algorithm simulates multiple partitions to determine that the entrance area X is expanded to 8 square meters, ensuring that the capacity matches the demand.

[0081] S1063: Based on the optimized division results, obtain the volume parameters of the goods to determine the preliminary layout scheme of the storage location; Based on the optimized partitioning results, cargo volume data is obtained to determine a preliminary allocation scheme for storage locations.

[0082] For example, B-002, with a volume of 0.5 cubic meters, is allocated to a forward position in zone X; C-003, with a volume of 1.2 cubic meters, has a large volume despite its low frequency, and is allocated to zone Y. This allocation takes into account space utilization efficiency.

[0083] S1064: The location parameters are integrated through the initial layout scheme, and the storage location is recalculated using the simulated annealing algorithm to obtain the adjusted layout scheme.

[0084] By integrating location parameters through the initial allocation scheme, the storage locations are recalculated using the simulated annealing algorithm to obtain the adjusted location layout.

[0085] Preferably, the location parameters include distance from the passageway; B-002 is adjusted to 1 meter from the entrance; and C-003 is moved to a position against the wall in area Y. The algorithm optimizes the spatial distribution of goods through multiple iterations, improving handling convenience.

[0086] In step S106, based on the storage location adjustment requirements, the storage locations are replanned by dividing the space into regions using a simulated annealing algorithm and combining it with cargo volume parameters to obtain an adjusted layout scheme. This is followed by the following steps: S1065: Analyze the matching degree of the coded associated data based on the adjusted layout scheme; S1066: Determine whether the matching degree is lower than the preset threshold. If the matching degree is lower than the preset threshold, readjust the layout information by dynamically storing data to determine the final optimized layout scheme.

[0087] The matching degree of the coded associated data is analyzed based on the adjusted position layout. If the matching degree is lower than the preset threshold, the layout information is readjusted by dynamically storing data.

[0088] For example, with a preset matching accuracy of 80%, the B-002 code indicates that it needs to be close to the forklift lane, but the current layout is far away, and the matching accuracy is only 60%.

[0089] In one embodiment, based on the latest inbound data, the system moves B-002 to the side of the forklift lane, increasing the matching accuracy to 85%, and determines the final layout. It then obtains the cargo identification and storage location data from the final optimized layout scheme to generate the adjusted complete layout plan.

[0090] Specifically, the plan outlines a complete mapping, such as A-001 in position 3 of zone Y and B-002 in position 1 of zone X. This plan ensures that the location of goods is highly aligned with dynamic needs, facilitating rapid response to operational requirements.

[0091] In step S107, based on the adjusted layout scheme, a pickup path is generated using the A* algorithm combined with path node density and real-time speed data. Simultaneously, path intersection probability and conflict range are detected to obtain a conflict-free optimal pickup path sequence. This includes the following steps: S1071: Obtain path node distribution data based on the adjusted layout plan, and integrate node density information to determine the initial pickup route sequence; Specifically, the path node distribution data is obtained through the adjusted layout scheme, and the node density information is integrated to determine the initial pickup route sequence.

[0092] Understandably, path node distribution data reflects the connection between the location of goods and the passageways in the warehouse.

[0093] For example, the adjusted layout shows that cargo B-002 is located at position 1 in zone X, near the entrance, while A-001 is located at position 3 in zone Y, towards the rear. Path nodes may be entrances, forklift lane turning points, and storage locations, while node density represents the number of nodes per unit area.

[0094] For example, zone X has a higher node density due to its proximity to the entrance passage, while zone Y has a lower density. The initial pickup route sequence might be from the entrance to position 1 in zone X, and then to position 3 in zone Y.

[0095] S1072: Analyze the traffic efficiency between path nodes using real-time speed data to obtain the dynamic adjustment requirements of the route sequence; By analyzing the traffic efficiency between path nodes using real-time speed data, the dynamic adjustment requirements of the route sequence can be obtained.

[0096] Specifically, real-time speed data comes from the actual operating records of forklifts or handling equipment.

[0097] For example, the speed from the entrance to position 1 in zone X is 2 meters per second because the distance is short and there are no obstacles. However, the speed from zone X to zone Y drops to 1 meter per second because the path is longer and may have corners. Based on this judgment, the dynamic adjustment requirement needs to shorten the low-speed section or optimize the inflection points.

[0098] In one possible implementation, the system detects low traffic efficiency from zone X to zone Y and suggests adjusting the route to avoid congested corners.

[0099] S1073: Detect the crossover probability in the route sequence based on dynamic adjustment requirements; S1074: Combining crossover probability and conflict range information, determine whether there is a conflict point. If a conflict point exists, regenerate the pickup route by adjusting the node density to obtain the updated route sequence. The system detects the intersection probability in the route sequence and, combined with the conflict range information, determines whether there are conflict points.

[0100] It should be noted that the crossover probability refers to the likelihood that multiple pickup routes overlap at a certain node.

[0101] For example, if two workers simultaneously pick up B-002 and C-003, their paths intersect in the forklift lane, resulting in a 70% probability of intersection. The conflict range information includes the width of the intersection; for example, if it is only 1.5 meters, the forklifts cannot run side by side.

[0102] In one embodiment, if a conflict point is detected, the system marks it as an area requiring optimization. If the conflict point exists, the pickup route is regenerated by adjusting the node density to obtain an updated route sequence.

[0103] Preferably, adjusting the node density can reduce the number of nodes at the intersection.

[0104] For example, some nodes in Zone X can be merged to reduce density, allowing routes to avoid intersections with forklift lanes. The updated route would lead directly from the entrance to position 1 in Zone X, and then along the side passage to position 3 in Zone Y. This adjustment reduces conflicts and improves traffic flow.

[0105] S1075: For the updated route sequence, obtain the real-time speed change trend of the path nodes to determine the preliminary plan for optimizing the route; For the updated route sequence, obtain the real-time speed change trend of the path nodes to determine the preliminary plan for optimizing the route.

[0106] Specifically, the trend of speed change can reflect the effect of node adjustment.

[0107] For example, the speed in the side passage stabilized at 1.8 m / s, which is better than the original corner section's 1 m / s. Therefore, the preliminary plan retains the side passage design to ensure improved efficiency.

[0108] S1076: The node density and speed data in the preliminary scheme of the optimization path are fused using the A* algorithm to calculate the optimal pickup path sequence without conflict.

[0109] The A* algorithm is used to fuse the node density and speed data in the preliminary scheme to calculate a conflict-free optimized path sequence.

[0110] In one embodiment, the A* algorithm uses node density as the weight and speed as the cost to calculate the path from the entrance to position 1 in zone X in just 3 seconds and to position 3 in zone Y in a total of 10 seconds.

[0111] For example, the optimized path sequence is entrance - side aisle - X zone position 1 - Y zone position 3, with no intersections, ensuring smooth handling. This method significantly improves the response speed of picking up goods and optimizes resource utilization.

[0112] In step S108, the handling equipment is guided to perform a picking operation based on the optimal picking path sequence, and real-time speed data and the location coordinates of dynamic obstacles are obtained from the operation log to determine the accuracy of the path intersection probability detection and update the conflict detection range. This includes the following steps: S1081: Drive the handling equipment to perform the picking task according to the optimal picking path sequence, and obtain real-time speed data and the position coordinates of dynamic obstacles from the operation log; Specifically, the optimal picking path sequence drives the handling equipment to perform picking tasks, and the operation log provides real-time speed data and the location coordinates of dynamic obstacles, which lays the foundation for understanding the task execution status.

[0113] Understandably, the operation log records the actual operating status of the handling equipment in the warehouse. For example, a forklift departs from the entrance and proceeds along the side aisle to pick up item B-002 at location 1 in zone X. Real-time speed data shows that it maintains a speed of 1.8 m / s in the side aisle, while the location coordinates of the dynamic obstacle show that there is a temporarily stacked item near location 2 in zone Y, with coordinates (5,12). These data reflect the operating characteristics in a dynamic environment.

[0114] S1082: Analyze the intersection probability between the location coordinates of dynamic obstacles and the picking path to determine potential conflict areas in the path sequence, and use real-time speed data to compare the passage efficiency of the handling equipment in the conflict area. Analyzing the location coordinates of dynamic obstacles and their probability of intersecting with the pickup path is crucial for identifying potential conflict areas.

[0115] Specifically, the forklift path is entrance - side aisle - X zone 1 position - Y zone 3 position, while the obstacle is located at (5,12), near the intersection of the Y zone aisles.

[0116] For example, if the obstacle occupies a width of 1 meter and the passage is only 2 meters wide, the forklift may be obstructed when passing through, and the probability of intersection can reach 60%.

[0117] In one possible implementation, the system calculates the intersection of the obstacle and the path based on coordinates, marking the area near position 2 in zone Y as a conflict zone. This analysis helps to identify risk points in advance. By comparing the passage efficiency of the transport equipment in the conflict zone using real-time speed data, the accuracy of the intersection probability detection can be further assessed.

[0118] For example, the forklift's speed was 1.8 m / s in the side passage, but dropped to 0.5 m / s when it approached position 2 in zone Y, indicating that the obstacle significantly affected passage.

[0119] Preferably, if the preset threshold is that the cross-probability detection accuracy needs to reach 80%, but the actual detection probability calculated based on speed changes is only 60%, then the detection is not accurate enough.

[0120] S1083: Determine whether the detection accuracy of the crossover probability meets the preset threshold. If the detection accuracy is lower than the preset threshold, adjust the conflict range by using the position coordinates of the dynamic obstacle to update the conflict detection range.

[0121] It should be noted that the correlation between the speed decrease and the obstacle's location directly affects the reliability of the judgment. If the detection accuracy is below the threshold, the conflict range is adjusted using the dynamic obstacle's position coordinates to provide updated detection data.

[0122] In one embodiment, the system expands the conflict range from a single point at position 2 in zone Y to a surrounding area of ​​1 meter, and adjusts the coordinate range to (4-6, 11-13).

[0123] For example, if the forklift slows down to 1 m / s at (5,11) in advance, it indicates that the obstacle's impact range exceeds expectations. After dynamic adjustment, the conflict area is closer to reality, and the detection criteria are optimized. This approach improves the adaptability of path planning, ensures that handling equipment avoids risky areas, and enhances the stability of task execution.

[0124] For example, the adjusted testing criteria can also support collaborative operation of multiple devices.

[0125] For example, if two forklifts are picking up goods at location 1 in zone X and location 3 in zone Y respectively, the system, based on the updated conflict range, plans for one forklift to go directly along the side passage while the other detours via the alternative route, avoiding congestion. This flexibility significantly enhances the scheduling efficiency of equipment within the warehouse.

[0126] In step S109, the path length change is analyzed based on the updated conflict detection range, and the storage layout is adjusted by using frequency fluctuations and dynamic obstacle coordinates to obtain an optimized resource allocation scheme.

[0127] By analyzing path length changes through conflict detection range, a weight exceedance state is obtained. If the weight exceedance state exceeds a preset threshold, obstacle coordinate data of dynamic obstacles is acquired to determine coordinate distribution characteristics. Based on the coordinate distribution characteristics, the frequency fluctuation is extracted to obtain the frequency fluctuation range. The storage layout structure is adjusted using the frequency fluctuation range to obtain the adjusted layout parameters. The resource allocation logic is updated using the adjusted layout parameters to obtain a preliminary allocation scheme. The preliminary allocation scheme is compared with the obstacle coordinate distribution to determine the degree of optimization. If the optimization degree is lower than a preset threshold, the final allocation scheme is obtained through iterative adjustment using the frequency fluctuation range and obstacle coordinates.

[0128] Specifically, path length changes are analyzed by examining the collision detection range.

[0129] Understandably, the core of this analysis lies in identifying the actual extension of the path caused by dynamic obstacles.

[0130] For example, a handling equipment was originally planned to have a path length of 20 meters from the warehouse entrance to position 1 in zone X. However, due to temporary storage of goods near position 2 in zone Y, the path was forced to detour to 25 meters. The increase in path length directly reflects the weight exceeding the state.

[0131] For example, if the preset threshold is that the path extension should not exceed 10%, i.e., 22 meters, but the actual extension exceeds 25 meters, then the determination of the weight exceeding the limit is triggered. This method intuitively reflects the impact of obstacles on the path.

[0132] In one possible implementation, once the weight exceeds a threshold, the system acquires the coordinate data of the dynamic obstacle.

[0133] For example, one obstacle is located at position 2 in zone Y, with coordinates 5, 12, and another obstacle is located at position 3 in zone Y, with coordinates 6, 14. The coordinate distribution characteristics show that the obstacles are concentrated near the passage in zone Y and are arranged linearly.

[0134] It should be noted that this distribution pattern suggests that obstacles may be formed due to the regular placement of temporary stockpiles. Therefore, extracting information on fluctuations in retrieval frequency is particularly important.

[0135] For example, the frequency of goods retrieval at location 1 in zone X was originally stable at 5 times per day, but it dropped to 3 times due to the detour. Meanwhile, the frequency at location 3 in zone Y fluctuated from 4 times to 2 times due to obstacles. The frequency fluctuation range was thus determined to be 2-5 times.

[0136] Specifically, the frequency fluctuation range can be used to adjust the storage layout structure.

[0137] In one embodiment, based on fluctuations of 2-5 times, the system moves high-frequency goods from position 2 in zone Y to a spare position in zone X, reducing the pressure on the zone Y channel. The adjusted layout parameters result in a 20% reduction in storage density in zone Y and a 15% increase in the utilization rate of the spare position in zone X.

[0138] Preferably, by updating the resource allocation logic using these parameters, the initial allocation scheme may deploy the two handling devices to Zone X and the backup channel respectively, to avoid congestion in Zone Y.

[0139] For example, one machine is responsible for retrieving goods from location 1 in zone X, while another machine travels along a backup route directly to location 3 in zone Y. For the initial allocation plan, it's necessary to compare the obstacle coordinate distribution to determine the degree of optimization.

[0140] For example, obstacle coordinates 5,12 and 6,14 indicate that there are still potential conflicts in the Y-zone channel. If the equipment detour time is reduced by only 10% after the solution is implemented, which is lower than the preset optimization threshold of 30%, then the optimization is insufficient.

[0141] In one possible implementation, the system iterates and adjusts the obstacle coordinates by frequency fluctuations ranging from 2 to 5 times.

[0142] For example, analysis revealed that the frequency decrease was mainly due to an obstacle at position 2 in zone Y. The cargo was then dispersed to the spare position at coordinates 4,11. The final allocation scheme was adjusted as follows: one machine retrieved cargo from zone X along the side passage, while the other machine detoured to position 3 in zone Y via the spare passage, reducing the detour time to 5%, close to the optimal state.

[0143] For example, this iterative adjustment can also support multi-device collaboration.

[0144] For example, when two handling devices pick up goods separately, the updated solution ensures that the paths do not intersect, and the equipment waiting time is reduced from 5 minutes to 2 minutes.

[0145] Understandably, the combination of frequency fluctuations and coordinate distribution not only optimizes individual paths but also improves overall scheduling efficiency. This approach provides a reliable basis for resource allocation in dynamic environments.

[0146] Reference Figure 2 The second embodiment of this application provides a smart factory warehouse management system based on big data, including: The preliminary location mapping module is used to assign an identification code to each item of goods using RFID technology, and to obtain the initial location data of the goods at the warehouse entrance by binding the goods to a data acquisition device equipped with sensors and recording it in the database, thereby obtaining the preliminary location mapping. The dynamic location distribution determination module is used to analyze the movement trajectory of the goods from the warehouse to the storage point by combining the positioning algorithm with multi-point sensor data based on the preliminary location mapping, so as to calculate the path node density and thus determine the current dynamic location distribution of the goods. The demand change trend quantification value acquisition module is used to calculate the frequency fluctuation of the pickup based on the dynamic location distribution and by using time series processing technology to fuse historical pickup data within the classification time window, thereby obtaining the quantification value of the demand change trend of the goods. The module for determining the optimal storage area range for high-frequency goods is used to determine whether the quantitative value of the demand change trend exceeds the retrieval frequency threshold. When the quantitative value of the demand change trend exceeds the retrieval frequency threshold, the module uses a clustering algorithm combined with retrieval frequency fluctuations and goods volume parameters to divide the goods into two categories: high-frequency and low-frequency, thereby determining the optimal storage area range for high-frequency goods. The storage location adjustment requirement acquisition module is used to obtain the code association data through the cargo identifier and integrate it into dynamic storage based on the optimized storage area range of the high-frequency goods, so as to obtain the storage location adjustment requirements. The adjusted layout scheme acquisition module is used to re-plan the storage location according to the adjustment requirements of the storage location by dividing the granularity by region using the simulated annealing algorithm and combining it with the cargo volume parameters, so as to obtain the adjusted layout scheme. The conflict-free optimal pickup path sequence acquisition module is used to generate pickup paths based on the adjusted layout scheme by combining the A* algorithm with path node density and real-time speed data, while detecting path intersection probability and conflict range to obtain a conflict-free optimal pickup path sequence; The conflict detection range update module is used to guide the handling equipment to perform the picking operation according to the optimal picking path sequence, and to obtain real-time speed data and the position coordinates of dynamic obstacles from the operation log to determine the accuracy of the path intersection probability detection and update the conflict detection range; The optimized resource allocation scheme acquisition module is used to analyze path length changes based on the updated conflict detection range, and adjust the storage layout by using frequency fluctuations and dynamic obstacle coordinates to obtain the optimized resource allocation scheme.

[0147] It should be noted that the smart factory warehouse management system based on big data provided in this embodiment of the invention is used to execute all process steps of the smart factory warehouse management method based on big data in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0148] This application also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a program for smart factory warehouse management based on big data. When the processor executes the computer program, it implements the steps described in the various embodiments of the smart factory warehouse management method based on big data, for example... Figure 1 The step S101 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the preliminary position mapping module.

[0149] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0150] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0151] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0152] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0153] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0154] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0155] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A smart factory warehouse management method based on big data, characterized in that, include: Each item is tagged with an RFID code, and the items are bound to a data acquisition device equipped with sensors to obtain initial location data of the items at the warehouse entrance and record it in the database, thereby obtaining a preliminary location mapping. Based on the preliminary location mapping, the movement trajectory of the goods from the warehouse to the storage point is analyzed by combining the positioning algorithm with multi-point sensor data to calculate the path node density, thereby determining the current dynamic location distribution of the goods. Based on the dynamic location distribution, historical pickup data within the classification time window are fused using time series processing technology to calculate the pickup frequency fluctuation, thereby obtaining a quantitative value of the demand change trend of the goods. The process involves determining whether the quantified value of the demand change trend exceeds a usage frequency threshold. If the quantified value exceeds the usage frequency threshold, a clustering algorithm is used in conjunction with usage frequency fluctuations and cargo volume parameters to classify the cargo into high-frequency and low-frequency categories, thereby determining the optimal storage area for high-frequency cargo. This includes: obtaining the quantified value of the demand change trend and usage frequency data from historical usage records, and determining whether the quantified value exceeds a preset usage frequency threshold to obtain an over-threshold marker; extracting frequency fluctuation data based on the over-threshold marker, and integrating it with the cargo volume parameters, then using a clustering algorithm to divide high-frequency and low-frequency cargo to determine the classification result; calculating the fluctuation amplitude of high-frequency cargo in the classification result through frequency fluctuation analysis to obtain a priority ranking of high-frequency cargo; and calculating the granularity of area division based on the priority ranking and the cargo volume parameters to determine the optimal storage area for high-frequency cargo. Based on the optimized storage area of ​​the high-frequency goods, the code association data is obtained through the goods identification and incorporated into dynamic storage to obtain the storage location adjustment requirements; Based on the storage location adjustment requirements, the storage locations are replanned using a simulated annealing algorithm with granular region division and cargo volume parameters to obtain an adjusted layout scheme. This includes: extracting the granular region division based on the storage location adjustment requirements and layout information; determining whether the granular region division meets the storage requirements; if the granular region division does not meet the storage requirements, adjusting the granular region division using a simulated annealing algorithm to obtain an optimized division result; obtaining the cargo volume parameters based on the optimized division result to determine a preliminary layout scheme for the storage locations; and integrating the location parameters with the preliminary layout scheme and recalculating the storage locations using a simulated annealing algorithm to obtain the adjusted layout scheme. Based on the adjusted layout scheme, the A* algorithm is used to generate pickup paths by combining path node density and real-time speed data collection. At the same time, the path intersection probability and conflict range are detected to obtain the optimal pickup path sequence without conflict. The optimal picking path sequence guides the handling equipment to perform picking operations, and real-time speed data and the location coordinates of dynamic obstacles are obtained from the operation log to determine the accuracy of path intersection probability detection and update the conflict detection range. The path length changes are analyzed based on the updated conflict detection range, and the storage layout is adjusted by using frequency fluctuations and dynamic obstacle coordinates to obtain an optimized resource allocation scheme.

2. The smart factory warehouse management method based on big data according to claim 1, characterized in that, The step involves analyzing the movement trajectory of the goods from entry into the warehouse to the storage point using a positioning algorithm combined with multi-point sensor data, based on the initial location mapping, to calculate the path node density and thus determine the current dynamic location distribution of the goods. This process then includes: The matching status between the location distribution of the goods and the storage points was verified through multi-point analysis. Determine whether there is a deviation between the location distribution and the storage points; if the deviation exists, cluster the sensor data using the K-means algorithm to obtain a set of abnormal nodes; Based on the set of abnormal nodes, the corresponding movement trajectory segments are extracted from the trajectory analysis to determine the time range of the abnormality. By using sensor data within the time range of the anomaly occurrence, combined with a positioning algorithm to correct the dynamic position, the adjusted position distribution of the goods can be obtained. Based on the adjusted location distribution, the current dynamic location distribution of the goods is updated by comparing the stored point data.

3. The smart factory warehouse management method based on big data according to claim 1, characterized in that, The step of fusing historical pickup data within a classification time window using time series processing techniques based on the dynamic location distribution to calculate pickup frequency fluctuations and thus obtain a quantitative value of the demand trend for the goods includes: Based on the dynamic location distribution, the storage location information of the goods is extracted to establish a location index table; The sliding time window method is used to obtain pickup records within a specified time range from the historical database, and the pickup frequency value corresponding to each location in the location index table is calculated by the frequency analysis algorithm. Determine whether the frequency value exceeds a preset threshold. If the frequency value exceeds the preset threshold, mark the location as a high-frequency access area. Based on the distribution of the high-frequency access areas, cluster analysis is used to divide the storage areas of the goods. By using time series processing technology, the frequency of access to each of the storage areas is analyzed for volatility. Based on the results of the volatility analysis, a regression model is used to quantify the demand change trend, so as to obtain a quantitative value of the demand change trend of the goods.

4. The smart factory warehouse management method based on big data according to claim 1, characterized in that, Based on the adjustment requirements of the storage location, the storage location is replanned by dividing it into regional granularities using a simulated annealing algorithm and combining it with cargo volume parameters to obtain an adjusted layout scheme, which then includes: Analyze the matching degree of the coded associated data based on the adjusted layout scheme; If the matching degree is lower than a preset threshold, the layout information is readjusted by dynamically storing data to determine the final optimized layout scheme.

5. The smart factory warehouse management method based on big data according to claim 1, characterized in that, The step of generating pickup routes based on the adjusted layout scheme, using the A* algorithm combined with path node density and real-time speed data, and simultaneously detecting path intersection probability and conflict range, to obtain a conflict-free optimal pickup route sequence includes: Based on the adjusted layout scheme, obtain path node distribution data and integrate node density information to determine the initial pickup route sequence; Real-time speed data is used to analyze the traffic efficiency between path nodes in order to obtain the dynamic adjustment requirements of the route sequence; Based on the dynamic adjustment requirements, detect the intersection probability in the route sequence; Combining the crossover probability and conflict range information, it is determined whether there is a conflict point. If the conflict point exists, the picking route is regenerated by adjusting the node density to obtain an updated route sequence. For the updated route sequence, the real-time speed change trend of the path nodes is obtained to determine a preliminary scheme for optimizing the path; The node density and speed data from the preliminary scheme of the optimized path are fused using the A* algorithm to calculate the conflict-free optimal pickup path sequence.

6. The smart factory warehouse management method based on big data according to claim 1, characterized in that, The step of guiding the handling equipment to perform picking operations according to the optimal picking path sequence, and obtaining real-time speed data and the position coordinates of dynamic obstacles from the operation log to determine the accuracy of path intersection probability detection and update the conflict detection range includes: The material handling equipment is driven to perform the material handling task according to the optimal material handling path sequence, and real-time speed data and the position coordinates of dynamic obstacles are obtained from the operation log; The probability of intersection between the location coordinates of the dynamic obstacle and the picking path is analyzed to determine the potential conflict area in the path sequence, and the passage efficiency of the handling equipment in the conflict area is compared with the real-time speed data. Determine whether the detection accuracy of the intersection probability meets a preset threshold. If the detection accuracy is lower than the preset threshold, adjust the conflict range by using the position coordinates of the dynamic obstacle to update the conflict detection range.

7. A storage medium, characterized in that, The storage medium stores computer program data, which, when executed, enables the implementation of the smart factory warehouse management method based on big data as described in any one of claims 1-6.

8. A smart factory warehouse management system based on big data, characterized in that, The method for implementing the big data-based smart factory warehouse management method as described in any one of claims 1 to 6 includes: The preliminary location mapping module is used to assign an identification code to each item of goods using RFID technology, and to obtain the initial location data of the goods at the warehouse entrance by binding the goods to a data acquisition device equipped with sensors and recording it in the database, thereby obtaining the preliminary location mapping. The dynamic location distribution determination module is used to analyze the movement trajectory of the goods from the warehouse to the storage point by combining the positioning algorithm with multi-point sensor data based on the preliminary location mapping, so as to calculate the path node density and thus determine the current dynamic location distribution of the goods. The demand change trend quantification value acquisition module is used to calculate the frequency fluctuation of the pickup based on the dynamic location distribution and by using time series processing technology to fuse historical pickup data within the classification time window, thereby obtaining the quantification value of the demand change trend of the goods. The module for determining the optimal storage area range for high-frequency goods is used to determine whether the quantified value of the demand change trend exceeds the retrieval frequency threshold. When the quantified value of the demand change trend exceeds the retrieval frequency threshold, the module uses a clustering algorithm combined with retrieval frequency fluctuations and goods volume parameters to classify the goods into high-frequency and low-frequency categories, thereby determining the optimal storage area range for high-frequency goods. Specifically, for high-frequency goods in the classification results, the module calculates the fluctuation amplitude through frequency fluctuation analysis to obtain a priority ranking of high-frequency goods; and calculates the granularity of area division by combining the priority ranking with the goods volume parameters to determine the optimal storage area range for high-frequency goods. The storage location adjustment requirement acquisition module is used to obtain the code association data through the cargo identifier and integrate it into dynamic storage based on the optimized storage area range of the high-frequency goods, so as to obtain the storage location adjustment requirements. The adjusted layout scheme acquisition module is used to re-plan the storage location according to the adjustment requirements of the storage location by dividing the granularity by region using the simulated annealing algorithm and combining it with the cargo volume parameters, so as to obtain the adjusted layout scheme. The conflict-free optimal pickup path sequence acquisition module is used to generate pickup paths based on the adjusted layout scheme by combining the A* algorithm with path node density and real-time speed data acquisition, and at the same time detect the path intersection probability and conflict range to obtain the conflict-free optimal pickup path sequence. The conflict detection range update module is used to guide the handling equipment to perform the picking operation according to the optimal picking path sequence, and to obtain real-time speed data and the position coordinates of dynamic obstacles from the operation log in order to determine the accuracy of the path intersection probability detection and update the conflict detection range. The optimized resource allocation scheme acquisition module is used to analyze path length changes based on the updated conflict detection range, and adjust the storage layout by using frequency fluctuations and dynamic obstacle coordinates to obtain the optimized resource allocation scheme.

Images