Automatic bag taking warehouse inventory real-time synchronization method and system

Abstract

This invention discloses a method and system for real-time inventory synchronization of an automatic bag-retrieving machine, relating to the field of inventory synchronization management. The method includes obtaining the bag types and quantities for the next bag-retrieving application cycle; performing matching verification during bag-retrieving tasks; merging high-frequency, continuous outbound events into batch events; and inputting inventory drift events into an edge computing unit to update the predicted additional bag-retrieving tasks for the next bag-retrieving application cycle. This invention, through a built-in edge computing unit, combines task data from the hospital information system with historical bag-retrieving data to dynamically generate internal tasks and predicted additional tasks for the next bag-retrieving cycle. During the outbound process, real-time matching verification is performed to prevent abnormal outbounds, identify inventory drift, and dynamically adjust predicted tasks, achieving a closed-loop inventory management system and reducing emergency allocation and operational risks. A cloud-based central dispatch platform optimizes allocation paths and resources, rapidly responds to emergency dispatch events, and enhances dispatch flexibility and the overall intelligence level of the system.

Description

Technical Field

[0001] This invention relates to the field of inventory synchronization management, specifically to a method and system for real-time inventory synchronization of an automatic bag-retrieving machine. Background Technology

[0002] Automated bag dispensing machines are crucial equipment in hospitals, pharmaceutical companies, and other settings for the automated management and intelligent dispensing of bags for medicines and consumables. The accuracy and real-time nature of their inventory management directly impacts the security and efficiency of material supply. Traditional automated bag dispensing machine inventory management typically relies on periodic manual inventory checks or simple inbound / outbound records, which suffers from problems such as delayed inventory data updates, cumbersome inventory checks, and difficulty in timely detection and handling of inventory anomalies. Furthermore, with the dynamic changes in hospital bag demand, especially during emergencies or peak periods, the problems of inventory shortages and untimely allocation responses become more prominent, easily leading to a shortage of clinical bags and affecting the quality of medical services.

[0003] Existing technology, such as the invention patent with announcement number CN112561449B, is a method and system for synchronizing inventory information. It includes a second server receiving an authentication request sent by a target user; the second server performing a first encryption process on the authentication request information using its private key, and then sending the first processing result to a first server; the first server performing a first decryption process on the first processing result using its public key to obtain a second processing result; when the second authentication request information matches the second processing result, the target user is confirmed to have passed authentication; when authentication is successful, the first server determines the updated data of the target entity to be synchronized based on the second authentication request information, and synchronizes the updated data of the target entity to be synchronized to a Redis server; the second server then reads the updated data information from the Redis server.

[0004] Existing technology, such as the invention patent with publication number CN104077671B, is a method and system for synchronizing inventory information. The method includes: dividing the total inventory information in the total inventory module into sub-inventory information corresponding to each sales module based on order information from each sales module, and setting an effective time for each sub-inventory information of each sales module; processing order requests based on the sub-inventory information of each sales module, sending the generated order information to the total inventory module, and synchronizing its corresponding sub-inventory information with the total inventory module according to the effective time of its own sub-inventory information; wherein, the total inventory module receives order information from each sales module and updates the total inventory information in the total inventory module according to the order information.

[0005] As can be seen from the above solutions, existing technologies in the field of inventory synchronization management typically rely on encrypted authentication and data synchronization between multiple servers. While this ensures security, it is not conducive to real-time updates and rapid responses to inventory status. Furthermore, existing technologies focus on data synchronization and authentication, which is insufficient to meet the complex needs of inventory changes in actual business operations. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for real-time inventory synchronization of an automatic bag-retrieving machine. To achieve the above objectives, this invention utilizes the following technical solution: A method for real-time inventory synchronization of an automatic bag-retrieving machine, comprising: The automatic bag-picking machine has a built-in edge computing unit that extracts HIS task data and historical bag-picking data. It generates internal bag-picking tasks and predicted additional bag-picking tasks for the next bag-picking application cycle, respectively. After summarizing these data, it obtains the bag types and quantities for the next bag-picking application cycle and sends them to the bag material warehouse system to apply for inventory for the next bag-picking application cycle.

[0007] During the bag retrieval application cycle, when the bag retrieval task is executed, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the erroneous bag retrieval is blocked. When the verification result is a match, the bag retrieval is executed, and the bag retrieval event data is generated and input into the processing engine.

[0008] Get the number of outbound events within a preset time window, merge high-frequency consecutive outbound events into batch events, and perform data compression and index storage.

[0009] The measured inventory is obtained based on RFID and sensor information, and the theoretical inventory is obtained based on bag retrieval task records. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds the threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag retrieval tasks for the next bag retrieval application cycle.

[0010] As a preferred technical solution, the process of generating internal bag-removing tasks and predicting additional bag-removing tasks for the next bag-removing application cycle is as follows: The automatic bag-retrieving machine's built-in edge computing unit extracts the HIS task data for the next bag-retrieving application cycle from the HIS. The HIS task data includes the bag type code, expected bag quantity, and expected bag-retrieving time for the area where the automatic bag-retrieving machine is located. This data is used to obtain the internal bag-retrieving task for the next bag-retrieving application cycle. The internal bag-retrieving task includes the bag type code, bag quantity, and expected outbound channel number. These data are then concatenated and combined according to a preset field order and data format to form a target triplet hash value.

[0011] The automatic bag-retrieving machine has a built-in edge computing unit that extracts historical bag-retrieving data, including the number of historical inventory drift events, the number of historical emergency dispatches, and the number of historical inventory shortages. Based on the historical bag-retrieving data, historical bag-retrieving feature values ​​are generated, and based on the historical bag-retrieving feature values, the additional bag-retrieving tasks for the next bag-retrieving application cycle are predicted.

[0012] The system summarizes the internal bag-retrieving tasks and predicted additional bag-retrieving tasks for the next bag-retrieving application cycle of the automatic bag-retrieving machine to obtain the bag types and quantities for the next bag-retrieving application cycle, and then sends the information to the bag material warehouse system to apply for inventory for the next bag-retrieving application cycle.

[0013] As a preferred technical solution, the edge computing unit performs matching and verification of the bag retrieval task. The specific process is as follows: For internal bag retrieval tasks, during the outbound execution phase, the built-in RFID reading module of the automatic bag retrieval machine uses high-frequency radio frequency identification to perform non-contact scanning of the bag box located at the outbound channel, collect the bag type code of the bag box in real time, and transmit the bag type code to the edge computing unit.

[0014] The edge computing unit extracts the bag type code from the currently executing bag retrieval task, and at the same time obtains the channel number of the currently executing outbound operation.

[0015] The edge computing unit concatenates and combines the bag type code, bag quantity, and outbound channel number of the bag box according to the preset field order and data format to form triplet index data, and caches the triplet data in a temporary verification queue.

[0016] The edge computing unit performs hash value calculation on the triple index data and matches it with the target triple hash value pre-generated and stored by the internal bag retrieval task. If the hash matching result does not match, the edge computing unit immediately sends a blocking command to the outbound execution module to control the outbound drive mechanism to stop its operation and trigger error event logging.

[0017] If the hash match is successful, the edge computing unit sends an execution instruction to the outbound execution module to drive the corresponding channel to complete the material outbound process. After the outbound process is completed, internal outbound event data containing task ID, bag type code, outbound channel number and outbound time is generated and fed into the processing engine in a streaming manner.

[0018] As a preferred technical solution, the edge computing unit performs matching and verification of the bag-retrieving task, and also includes: If it is an additional bag retrieval task, the edge computing unit calculates the available predicted reserve of the bag type in the current bag retrieval application cycle and matches it with the target bag retrieval quantity of the additional bag retrieval task. If the available predicted reserve is insufficient, an inventory shortage reminder is sent to the automatic bag retrieval machine display terminal, and the data of the additional bag retrieval task is uploaded to the cloud central scheduling platform to generate an emergency scheduling event.

[0019] If there is sufficient available forecast margin, the material will be discharged from the warehouse, driving the corresponding discharge channel to complete the discharge of the bagged material, generating additional discharge event data, and inputting the additional discharge event data into the processing engine in a streaming manner.

[0020] As a preferred technical solution, the number of outbound events within a preset time window is obtained, high-frequency consecutive outbound events are merged into batch events, and data compression and index storage are performed, specifically including: The edge computing unit continuously monitors the arrival timestamps of events from the streaming data stream of outbound events, based on a sliding time window mechanism, and counts the number of outbound events for various types of bag codes within a preset time window.

[0021] When the number of times a certain type of bagged material leaves the warehouse exceeds a preset high-frequency threshold within a continuous time window, adjacent events within that time window are merged to generate a single batch event record. The batch event records the bagged material type code, batch quantity, corresponding task ID sequence, and set of warehouse exit channel numbers.

[0022] After generating batch events, the processing engine performs structured compression on the event data, replacing repetitive fixed fields with a single-time storage method, and saving dynamic fields in differential encoding form.

[0023] As a preferred technical solution, the measured inventory is obtained based on RFID and sensor information, while the theoretical inventory is obtained based on bag-retrieving task records, specifically including: During each bag-picking application cycle, the edge computing unit collects in real time the unique identifier of the bag box identified by the built-in RFID module of the automatic bag picking machine and the occupancy status of the bag box in the warehouse detected by the sensor, and calculates the actual inventory quantity of various types of bags.

[0024] Meanwhile, based on the bag retrieval task records within the bag retrieval application cycle, including the execution records of internal bag retrieval tasks and predicted additional bag retrieval tasks, the theoretical inventory quantity of the corresponding bag material type code is calculated by accumulating the number of tasks.

[0025] As a preferred technical solution, the edge computing unit compares the measured inventory with the theoretical inventory, specifically including: The edge computing unit performs a difference comparison between the measured inventory quantity and the theoretical inventory quantity, calculates the inventory deviation value, and if the absolute value of the inventory deviation value exceeds the preset inventory deviation threshold, an inventory drift event is immediately generated. The inventory drift event includes the bag type code, measured inventory quantity, theoretical inventory quantity, inventory deviation value, trigger time, and trigger reason code.

[0026] The inventory drift event is input into the edge computing unit to update the predicted additional bag retrieval task for the next bag retrieval application cycle. Specifically, the original predicted additional bag retrieval quantity is corrected according to the deviation direction. The correction result is used to update the predicted additional bag retrieval task for the next cycle and synchronized to the bag material warehouse system and the cloud central scheduling platform for inventory allocation and transfer calculation.

[0027] As a preferred technical solution, an emergency dispatch event is generated, and the specific processing conditions are as follows: After an emergency dispatch event is generated, the edge computing unit uploads it to the central dispatch platform in the cloud.

[0028] Upon receiving an emergency dispatch event, the cloud-based central dispatch platform obtains real-time inventory snapshots of other automatic bag-retrieving machines and information on remaining bag-retrieving tasks within the current bag-retrieving application cycle, and calculates the available capacity of each other automatic bag-retrieving machine.

[0029] After the adjustable reserve is calculated, the cloud-based central dispatch platform optimizes the allocation by taking into account the geographical distance between other automatic bag-retrieving machines and event-triggered devices, the allocation time, and the sufficiency of the reserve, generates temporary allocation tasks, and sends them to the support automatic bag-retrieving machines for execution.

[0030] As a preferred technical solution, the generation of emergency scheduling events also includes generating an emergency scheduling event by the edge computing unit if the total output of a certain type of bag material from the automatic bag-retrieving machine exceeds a sudden threshold. The specific processing conditions are as follows: During the execution of the bag retrieval application cycle, when the cumulative outbound quantity of a certain bag material type code exceeds the preset sudden threshold within the preset bag retrieval time window, the edge computing unit immediately generates an emergency scheduling event.

[0031] After an emergency dispatch event is generated, the edge computing unit immediately performs local processing, pushes the emergency bag usage warning information to the automatic bag dispensing machine display terminal, and uploads it to the cloud central dispatch platform for allocation and optimization processing.

[0032] Simultaneously, extract the remaining internal bag retrieval tasks within the current bag retrieval application cycle, obtain the required bag types and quantities for the remaining internal bag retrieval tasks, compare them with the local inventory, and if the local inventory can meet the required bag types and quantities for the remaining internal bag retrieval tasks, obtain the inventory balance between the required bag types and quantities for the remaining internal bag retrieval tasks and the local inventory, and send the inventory balance to the cloud central scheduling platform to prioritize the allocation optimization.

[0033] If the local inventory cannot meet the required bag types and quantities for the remaining internal bag retrieval tasks, obtain the inventory shortage between the required bag types and quantities for the remaining internal bag retrieval tasks and the local inventory, and send the inventory shortage to the cloud central scheduling platform to prioritize the allocation optimization.

[0034] An automatic bag-retrieving machine inventory real-time synchronization system includes: The prediction module is used to extract HIS task data and historical bag retrieval data, generate internal bag retrieval tasks and predict additional bag retrieval tasks for the next bag retrieval application cycle, summarize them to obtain the bag types and quantities for the next bag retrieval application cycle, and then send them to the bag material warehouse system to apply for inventory for the next bag retrieval application cycle.

[0035] The matching module is used in the bag retrieval application cycle. When the bag retrieval task is executed, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the incorrect bag retrieval is blocked. When the verification result is a match, the bag retrieval is executed, and the bag retrieval event data is generated and input into the processing engine.

[0036] The compression module is used to obtain the number of outbound events within a preset time window, merge high-frequency continuous outbound events into batch events, and perform data compression and index storage.

[0037] The update module is used to obtain the measured inventory based on RFID and sensor information, and the theoretical inventory based on the bag-retrieving task records. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds a threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag-retrieving tasks for the next bag-retrieving application cycle. Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects: (1) This invention provides a method for real-time inventory synchronization of an automatic bag-retrieving machine. By using a built-in edge computing unit, it extracts task data and historical bag-retrieving data from the Hospital Information System (HIS) in real time, combines internal bag-retrieving tasks with predicted additional bag-retrieving tasks, and dynamically summarizes the demand for the next bag-retrieving application cycle. This data-driven inventory forecasting and application method effectively improves the accuracy of inventory applications, reduces inventory backlog and shortages, ensures timely fulfillment of clinical needs for medical bags, and enhances the intelligent level of hospital material management.

[0038] (2) In the outbound execution phase, this invention employs a matching verification mechanism based on RFID and sensors to strictly verify the bag-picking task and prevent erroneous outbound events. By blocking abnormal operations in real time and generating outbound event data promptly, the accuracy and traceability of material outbound are ensured. At the same time, batch event merging and data compression technologies improve the efficiency of event data processing, reduce system storage and transmission pressure, and ensure the efficient and stable operation of the system.

[0039] (3) This invention achieves timely detection and feedback of inventory drift events by comparing the difference between measured inventory and theoretical inventory. By dynamically correcting the predicted additional bag-taking tasks through inventory drift events, a closed-loop inventory control mechanism is formed, which improves the accuracy and responsiveness of inventory forecasting, reduces emergency allocation caused by inventory deviation, and lowers operational risks and costs.

[0040] (4) This invention establishes a collaborative allocation mechanism between a cloud-based central dispatch platform and an automatic bag-retrieving machine. In the event of an emergency dispatch, the platform comprehensively considers geographical distance, allocation time, and sufficiency of reserves to optimize allocation paths and resource allocation, thereby improving emergency response speed and allocation efficiency. At the same time, based on a priority adjustment strategy for local inventory, it achieves reasonable allocation and priority management of allocation resources, enhancing the system's flexible scheduling capabilities and the overall intelligence level of inventory management.

[0041] Of course, any product implementing this invention does not necessarily need to achieve all of the above advantages at the same time. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0043] Figure 2 This is a schematic diagram of the system modules of the present invention.

[0044] Figure 3 This is a schematic diagram of the logic flow of the present invention.

[0045] Figure 4 This is a schematic diagram of the task generation and warehouse verification logic flow of the present invention.

[0046] Figure 5 This is a schematic diagram of the event statistics and inventory monitoring logic flow of the present invention.

[0047] Figure 6 This is a schematic diagram of the emergency dispatch and support execution logic flow of the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] In the description of this invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inner", "around", etc., which indicate orientation or positional relationship, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting this invention.

[0050] Please see Figure 1 As shown, this embodiment of the invention provides a method for real-time inventory synchronization of an automatic bag-retrieving machine, specifically including: like Figure 3 The diagram shown illustrates the logical flow involved in this embodiment of the invention. The entire process of real-time inventory synchronization and emergency allocation for the automatic bag-picking machine is presented visually, making the work sequence, condition judgments, and data flow paths more intuitive. From extracting HIS data and historical bag-picking data, generating internal and predictive tasks, and applying for inventory for the next cycle, to outbound execution, inventory discrepancy detection, triggering emergency scheduling, and cloud optimization, the steps are presented in a complete logical sequence, avoiding omissions of key links. In nodes such as outbound matching verification, inventory discrepancy comparison, and sudden threshold judgment, the processing methods under different conditions are intuitively displayed through visual branch paths, facilitating quick understanding of the logic by R&D, testing, and maintenance personnel. Simultaneously, the interaction between data flow and control flow is intuitively demonstrated, showcasing the data transmission and instruction issuance process between the edge computing unit, the bag warehouse system, and the cloud-based central scheduling platform.

[0051] Figure 4 This is a schematic diagram of the task generation and outbound verification logic flow of the present invention. It mainly demonstrates how the automatic bag-retrieving machine generates internal bag-retrieving tasks and predicts additional bag-retrieving tasks within the edge computing unit, and performs verification before outbound processing. By comparing the triple hashes of the tasks, the accuracy and security of each outbound operation are ensured, preventing erroneous outbound processing or task tampering.

[0052] The automatic bag-retrieving machine has a built-in edge computing unit that extracts task data and historical bag-retrieving data from the HIS (Hospital Information System). It then generates internal bag-retrieving tasks and predicted additional bag-retrieving tasks for the next bag-retrieving application cycle. After summarizing these data, it obtains the bag types and quantities for the next bag-retrieving application cycle and sends them to the bag material warehouse system to request inventory for the next bag-retrieving application cycle.

[0053] The automatic bag-retrieving machine has a built-in edge computing unit that extracts HIS task data for the next bag-retrieving application cycle from the HIS. In this embodiment of the invention, the built-in edge computing unit of the automatic bag-retrieving machine refers to the edge computing module integrated inside the automatic bag-retrieving machine, which is responsible for local data processing, task generation and inventory management, reducing dependence on the cloud and achieving rapid response and real-time control.

[0054] HIS task data includes the bag type code, expected bag quantity, and expected bag retrieval time for the area where the automatic bag retrieval machine is located. This data is used to obtain the internal bag retrieval task for the next application cycle. The internal bag retrieval task includes the bag type code, bag quantity, and expected exit channel number. These are concatenated according to a preset field order and data format to form a target triplet hash value. Specifically, the edge computing unit extracts key fields from the internal bag retrieval task sequentially according to a predefined field order. In this embodiment, these are the bag type code, bag quantity, and expected exit channel number. These fields are then seamlessly concatenated into a continuous data string, avoiding unnecessary separators or spaces. After concatenation, the edge computing unit applies a preset hash algorithm (e.g., SHA-256 or MD5) to this string to calculate a fixed-length hash value. This hash value serves as a unique identifier for the target triplet, used for rapid comparison and task verification, ensuring that task information is not tampered with or misused during transmission and execution.

[0055] By seamlessly concatenating key fields in the internal bag-retrieving task according to a preset field order and data format, and applying a hash algorithm to generate a fixed-length target triplet hash value, the security and efficiency of task verification can be significantly improved. This hash value, serving as a unique digital fingerprint for the task, effectively prevents data from being tampered with or misused during transmission or execution, as any field change will result in a completely different hash value, thus allowing for immediate identification. During the outbound execution phase, only a single equality comparison between the real-time generated triplet hash value and the pre-stored target hash value is needed to complete the task matching verification, avoiding the complex calculations of field-by-field comparison and significantly improving verification speed. Simultaneously, this mechanism ensures that only completely matching tasks are allowed to proceed, preventing operational errors such as incorrect bags, incorrect quantities, and incorrect channels from the source, reducing inventory drift risks, and achieving high-speed, accurate, and secure execution of bag-retrieving tasks in an edge computing environment.

[0056] The bag-taking application cycle is a fixed time window, which in this embodiment of the invention is a time window of 1 day, used for planning and scheduling inventory tasks.

[0057] The automatic bag-retrieving machine has a built-in edge computing unit that extracts historical bag-retrieving data. Specifically, this extraction is accomplished by querying corresponding tables in a structured database. The historical bag-retrieving data includes the number of historical inventory drift events, the number of historical emergency dispatches, and the number of historical inventory shortages. Based on this historical bag-retrieving data, historical bag-retrieving feature values ​​are generated. The specific process includes: The database extracts a set of historical bag-retrieving data verification values, including verification values ​​for historical inventory drift events, historical emergency dispatch events, and historical inventory shortage events. The historical inventory drift event verification value refers to the total number of inventory drift events detected by the automatic bag-retrieving machine within a specified time range (e.g., the most recent N bag-retrieving application cycles). An inventory drift event is an abnormal event where the absolute value of the deviation between the measured inventory and the theoretical inventory exceeds a preset threshold. This number reflects the frequency of errors or abnormal fluctuations in inventory management. The historical emergency dispatch verification value refers to the total number of emergency dispatch events triggered by the automatic bag-retrieving machine due to inventory shortages or sudden demand within a specified time range (e.g., the most recent N bag-retrieving application cycles). This number characterizes the temporary replenishment scheduling pressure faced by the equipment in the past, reflecting the urgency and frequency of inventory replenishment. The historical inventory shortage verification value refers to the number of inventory shortage alarms detected by the automatic bag-retrieving machine within a specified time, i.e., the number of alarm events triggered when the inventory is lower than the preset safety stock level. This number indicates the frequency of inventory safety risks. The historical bag-retrieving data of the automatic bag-retrieving machine is compared with the historical bag-retrieving data verification value set, and then weighted and summed to obtain the historical bag-retrieving feature value of the automatic bag-retrieving machine. The specific calculation formula is as follows: ; in, Here, py represents the historical bag-retrieving characteristic value of the automatic bag-retrieving machine, td represents the historical inventory drift event count of the automatic bag-retrieving machine, kr represents the historical inventory shortage count of the automatic bag-retrieving machine, py0 represents the historical inventory drift event count verification value, td0 represents the historical emergency dispatch count verification value, and kr0 represents the historical inventory shortage count verification value. As a weighting factor for the number of historical inventory drift events, As a weighting factor for the number of historical emergency dispatches, Weighting factor for the number of times historical inventory was insufficient.

[0058] It should be noted that the weighting factors for historical inventory drift events, historical emergency dispatch events, and historical inventory shortages are as follows: the weighting factor for historical inventory drift events is used to adjust the weight of the impact of inventory drift events on the forecast; the weighting factor for historical emergency dispatch events reflects the contribution weight of emergency dispatch events to the forecast; and the weighting factor for historical inventory shortages represents the weight of inventory shortages on the forecast. In this embodiment of the invention, the weighting factors are obtained and dynamically adjusted through machine learning to adapt to changes in different time periods and actual inventory conditions.

[0059] It's also important to note that there's a correlation between the parameters of historical inventory drift events, historical emergency dispatch events, and historical inventory shortages. The number of historical inventory drift events reflects the frequency and magnitude of the deviation between theoretical and actual inventory levels; this deviation is often one of the root causes of inventory anomalies. Frequent inventory drift indicates compromised inventory data accuracy, potentially leading to abnormal execution of subsequent outbound plans. Historical inventory shortages are often a direct consequence of inventory drift. Inventory drift results in a lower-than-expected quantity of bags available in the system, causing inventory shortages and hindering normal bag-picking operations, impacting bag-using efficiency and business continuity. The number of historical emergency dispatch events reflects the frequency of emergency allocation actions taken by the system to address inventory shortages or sudden bag-using demands. Emergency dispatch is typically a remedial measure initiated to quickly replenish inventory after inventory drift or shortages occur.

[0060] The prediction of additional bag retrieval tasks for the next bag retrieval application cycle based on historical bag retrieval feature values ​​specifically includes inputting historical bag retrieval feature values ​​into the database pre-stored historical bag retrieval feature values ​​minus the predicted additional bag retrieval quantity replenishment ratio, obtaining the predicted additional bag retrieval quantity replenishment ratio for each type of bag material, multiplying the predicted additional bag retrieval quantity replenishment ratio for each type of bag material by the quantity of each type of bag material summarized in the internal bag retrieval task, and obtaining the prediction of additional bag retrieval tasks for the next bag retrieval application cycle, including the predicted additional bag material type code, the additional bag retrieval quantity, and the expected outbound channel number.

[0061] It should be noted that, in this embodiment of the invention, the edge computing unit pre-maintains a channel resource pool, including outbound channel numbers and status information. When an additional bag retrieval task is predicted, the edge computing unit executes a channel allocation algorithm based on the status information of the channel resource pool to sequentially obtain the expected outbound channel numbers for the predicted additional replacement tasks. In this embodiment, the channel allocation algorithm is the shortest queue first algorithm. For each outbound channel, the number of tasks currently waiting in the queue is counted, and the channel with the shortest queue is selected first.

[0062] The system summarizes the internal bag-retrieving tasks and predicted additional bag-retrieving tasks for the next bag-retrieving application cycle of the automatic bag-retrieving machine to obtain the bag types and quantities for the next bag-retrieving application cycle, and then sends the information to the bag material warehouse system to apply for inventory for the next bag-retrieving application cycle.

[0063] During the bag retrieval application cycle, when the bag retrieval task is executed, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the erroneous bag retrieval is blocked. When the verification result is a match, the bag retrieval is executed, and the bag retrieval event data is generated and input into the processing engine.

[0064] For internal bag retrieval tasks, during the outbound execution phase, the built-in RFID reading module of the automatic bag retrieval machine uses high-frequency radio frequency identification to perform non-contact scanning of the bag box located at the outbound channel, collect the bag type code of the bag box in real time, and transmit the bag type code to the edge computing unit.

[0065] The edge computing unit extracts the bag type code from the currently executing bag retrieval task, and at the same time obtains the channel number of the currently executing outbound operation.

[0066] The edge computing unit concatenates and combines the bag type code, bag quantity, and outbound channel number of the bag box according to the preset field order and data format to form triplet index data, and caches the triplet data in a temporary verification queue.

[0067] The edge computing unit performs hash value calculation on the triple index data and matches it with the target triple hash value pre-generated and stored by the internal bag retrieval task. If the hash matching result does not match, the edge computing unit immediately sends a blocking command to the outbound execution module to control the outbound drive mechanism to stop its operation and trigger error event logging.

[0068] If the hash match is successful, the edge computing unit sends an execution instruction to the outbound execution module to drive the corresponding channel to complete the material outbound process. After the outbound process is completed, internal outbound event data containing task ID, bag type code, outbound channel number and outbound time is generated and fed into the processing engine in a streaming manner.

[0069] If it is an additional bag retrieval task, the edge computing unit calculates the available predicted reserve of the bag type in the current bag retrieval application cycle and compares it with the target bag retrieval quantity of the additional bag retrieval task. If the available predicted reserve is insufficient, an inventory shortage reminder is sent to the automatic bag retrieval machine display terminal, and the data of the additional bag retrieval task is uploaded to the cloud central scheduling platform to generate an emergency scheduling event.

[0070] like Figure 6 The diagram illustrates the emergency dispatching and support execution logic of this invention. It shows how, when the total outbound volume exceeds the emergency threshold, the edge computing unit generates an emergency dispatching event and uploads it to the cloud. The cloud-based central dispatching platform calculates and optimizes the allocation of support automatic bag-retrieving machines, and issues support tasks. Finally, the support automatic bag-retrieving machines execute the outbound dispatching.

[0071] After an emergency dispatch event is generated, the edge computing unit uploads it to the central dispatch platform in the cloud.

[0072] Upon receiving an emergency dispatch event, the cloud-based central dispatch platform obtains real-time inventory snapshots of other automatic bag-retrieving machines and information on remaining bag-retrieving tasks within the current bag-retrieving application cycle, and calculates the available capacity of each other automatic bag-retrieving machine.

[0073] After the adjustable reserve is calculated, the cloud-based central dispatch platform optimizes the allocation by considering the geographical distance between other automatic bag-retrieving machines and event-triggered devices, the allocation time, and the sufficiency of the reserve. It then generates temporary allocation tasks and distributes them to the support automatic bag-retrieving machines for execution. Specifically, this includes: The central dispatch platform calculates the distance between the automated bag-retrieving machine and the event-triggered device based on their preset geographical locations. The allocation time is calculated based on historical allocation records. The cloud-based central dispatch platform calculates the remaining available quantity (i.e., allocateable surplus) of the required bag types in emergency dispatch events based on the current inventory snapshot of each automated bag-retrieving machine, and assesses the sufficiency of the surplus by combining this with the remaining internal bag-retrieving tasks in the current bag-retrieving application cycle. This metric is expressed as a percentage.

[0074] The geographical distances between each other automatic bag-retrieving machine and the event-triggered device are input into the pre-stored mapping set of geographical distance-allocation geographical optimization factor in the database. The mapping and matching are then performed to obtain the allocation geographical optimization factor for each other automatic bag-retrieving machine. The smaller the geographical distance, the larger the allocation geographical optimization factor.

[0075] The allocation time of each other automatic bag-retrieving machine is entered into the pre-stored mapping set of allocation time-allocation time optimization factor in the database and mapped and matched to obtain the allocation time optimization factor of each other automatic bag-retrieving machine. The shorter the allocation time, the larger the allocation time optimization factor.

[0076] The sufficiency of the remaining balance of each other automatic bag-retrieving machine is entered into the pre-stored mapping set of sufficiency of remaining balance and allocation of remaining balance optimization factor in the database. The mapping and matching are performed to obtain the allocation of remaining balance optimization factor of each other automatic bag-retrieving machine. The larger the sufficiency of remaining balance, the larger the allocation of remaining balance optimization factor.

[0077] Obtain the sum of the allocation geographical optimization factor, allocation geographical optimization factor, and allocation surplus optimization factor of each other automatic bag-retrieving machine, and denot it as the allocation optimization factor. Based on the magnitude of the allocation optimization factor, sort each other automatic bag-retrieving machine from high to low to form a candidate support automatic bag-retrieving machine sequence. Based on the required quantity of bag types in the emergency dispatch event, select support automatic bag-retrieving machines that meet the required quantity of bag types from the sequence in turn.

[0078] The automatic bag-retrieving machine is a device selected by a cloud-based central dispatch platform to provide bag material support to other automatic bag-retrieving machines in need during emergency dispatch or when inventory is insufficient.

[0079] After receiving a support task from the central cloud-based dispatch platform, the automatic support bag retrieval machine analyzes the task to determine the type, quantity, and target exit channel of the bags to be allocated. The edge computing unit of the automatic support bag retrieval machine then inputs the support task information into the local dispatch system, generates a corresponding support bag retrieval task, and arranges the material exit according to priority and workflow.

[0080] The generation of emergency scheduling events also includes the following: if the total output of a certain type of bag material from the automatic bag-retrieving machine exceeds a sudden threshold, the edge computing unit generates an emergency scheduling event. The specific processing conditions are as follows: It's important to note that the scheduling optimization strategy in the cloud-based central dispatch platform relies on real-time synchronization of inventory snapshots. The platform collects and updates inventory information for each automated bag-retrieving machine in real time, including bag type, remaining quantity, and task completion status, forming a dynamic inventory database to provide accurate data support for subsequent scheduling. The platform comprehensively analyzes current bag demand, dynamically assessing the urgency and priority of each node, prioritizing automated bag-retrieving machines with tight inventory and urgent needs. By dynamically assessing the urgency and priority of each node, the platform employs a multi-objective optimization algorithm to allocate tasks reasonably. Simultaneously, the platform uses a dynamic priority weight adjustment mechanism to adjust the priority of allocation tasks in real time based on local inventory conditions, ensuring flexible allocation decisions that meet actual business needs and effectively respond to sudden bag demand.

[0081] During the execution of the bag retrieval application cycle, when the cumulative outbound quantity of a certain bag material type code exceeds the preset sudden threshold within the preset bag retrieval time window, the edge computing unit immediately generates an emergency scheduling event.

[0082] After an emergency dispatch event is generated, the edge computing unit immediately performs local processing, pushes the emergency bag usage warning information to the automatic bag dispensing machine display terminal, and uploads it to the cloud central dispatch platform for allocation and optimization processing.

[0083] Simultaneously, the remaining internal bag-retrieving tasks within the current bag-retrieving application cycle are extracted to obtain the required bag types and quantities for the remaining internal bag-retrieving tasks. This is compared with the local inventory. If the local inventory can meet the required bag types and quantities for the remaining internal bag-retrieving tasks, the difference between the local inventory and the required bag types and quantities for the remaining internal bag-retrieving tasks is calculated to obtain the inventory surplus. The inventory surplus is then sent to the cloud central scheduling platform for priority downgrading of allocation optimization. Specifically, the inventory surplus is input into the pre-stored inventory surplus-priority weight downgrading factor mapping set in the database for mapping and matching to obtain the priority weight downgrading factor. This factor is then entered into the cloud central scheduling platform and multiplied with the original priority weight to obtain the downgraded priority weight.

[0084] If the local inventory cannot meet the needs of the remaining internal bag retrieval tasks in terms of bag types and quantities, the difference between the required bag types and quantities for the remaining internal bag retrieval tasks and the local inventory is used to obtain the inventory shortage. The inventory shortage is then sent to the cloud central scheduling platform for priority adjustment of allocation optimization. Specifically, the inventory shortage is entered into the pre-stored mapping set of inventory shortage-priority weight adjustment factor in the database, and the mapping is matched to obtain the priority weight adjustment factor. This factor is then entered into the cloud central scheduling platform and multiplied with the original priority weight to obtain the adjusted priority weight.

[0085] If there is sufficient available forecast margin, the material will be discharged from the warehouse, driving the corresponding discharge channel to complete the discharge of the bagged material, generating additional discharge event data, and inputting the additional discharge event data into the processing engine in a streaming manner.

[0086] Get the number of outbound events within a preset time window, merge high-frequency consecutive outbound events into batch events, and perform data compression and index storage.

[0087] The edge computing unit continuously monitors the arrival timestamps of events from the streaming data stream of outbound events, based on a sliding time window mechanism, and counts the number of outbound events for various types of bag codes within a preset time window.

[0088] When the number of times a certain type of bagged material leaves the warehouse exceeds a preset high-frequency threshold within a continuous time window, adjacent events within that time window are merged to generate a single batch event record. The batch event records the bagged material type code, batch quantity, corresponding task ID sequence, and set of warehouse exit channel numbers.

[0089] After generating batch events, the processing engine performs structured compression on the event data, replacing repetitive fixed fields with a single-time storage method, and saving dynamic fields in differential encoding form.

[0090] Grouping high-frequency, continuous outbound events into batch events significantly reduces the processing and storage pressure on edge computing units and backend systems, thus reducing the number of events. Through structured compression, repetitive fixed fields only need to be stored once, while dynamic fields are saved using differential encoding, effectively saving storage space. The reduction in data volume also improves data transmission efficiency and reduces network bandwidth consumption, making it particularly suitable for edge-cloud environments with limited network conditions. Simultaneously, this batch processing method reduces the frequent processing of individual events, improving the real-time response capabilities for inventory synchronization and anomaly detection. Furthermore, batch event records contain complete task ID sequences and outbound channel sets, preserving necessary details for subsequent traceability and refined inventory management. Overall, this method significantly improves system performance and resource utilization efficiency while ensuring data integrity and traceability.

[0091] like Figure 5 The diagram illustrates the event statistics and inventory monitoring logic flow of this invention. It demonstrates how the edge computing unit statistically analyzes outbound events, merges high-frequency continuous events to generate batch records, compares actual inventory with theoretical inventory, generates inventory drift events, and updates the prediction task.

[0092] The measured inventory is obtained based on RFID and sensor information, and the theoretical inventory is obtained based on bag retrieval task records. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds the threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag retrieval tasks for the next bag retrieval application cycle.

[0093] During each bag-retrieving application cycle, the edge computing unit collects in real time the unique identifier of the bag box identified by the built-in RFID module of the automatic bag-retrieving machine and the occupancy status of the bag box in the storage space detected by the sensor. It then calculates the actual inventory quantity of various types of bags. Specifically, in this embodiment, the automatic bag-retrieving machine is locally deployed with storage space sensors. The sensors detect the occupancy status of each storage location and determine whether a bag box is present in that location. All bag box information that is occupied and successfully identified by RFID is summarized and categorized, grouped according to bag type, to finally obtain the actual inventory quantity corresponding to each type of bag.

[0094] It's important to note that using occupancy sensors to confirm the actual occupancy of storage spaces effectively avoids misjudgments in inventory due to RFID tag obstruction, damage, or abnormal reading / writing. This ensures that only bags that are actually present and successfully identified are counted in inventory. Simultaneously, combining this with unique RFID identifiers enables precise classification of bag types, reflecting detailed changes in inventory structure in real time and supporting refined management. This dual-confirmation mechanism not only enhances the reliability of inventory data but also improves the inventory monitoring efficiency of the automated bag-picking machine, guaranteeing the accuracy of subsequent bag-picking tasks and the stability of inventory synchronization.

[0095] Meanwhile, based on the bag retrieval task records within the bag retrieval application cycle, including the execution records of internal bag retrieval tasks and predicted additional bag retrieval tasks, the theoretical inventory quantity of the corresponding bag material type code is calculated by accumulating the number of tasks.

[0096] The edge computing unit performs a difference comparison between the measured inventory quantity and the theoretical inventory quantity, calculates the inventory deviation value, and if the absolute value of the inventory deviation value exceeds the preset inventory deviation threshold, an inventory drift event is immediately generated. The inventory drift event includes the bag type code, measured inventory quantity, theoretical inventory quantity, inventory deviation value, trigger time, and trigger reason code.

[0097] The inventory drift event is input into the edge computing unit to update the predicted additional bag retrieval task for the next bag retrieval application cycle. Specifically, the original predicted additional bag retrieval quantity is corrected according to the deviation direction. The correction result is used to update the predicted additional bag retrieval task for the next cycle and synchronized to the bag material warehouse system and the cloud central scheduling platform for inventory allocation and transfer calculation.

[0098] Based on the direction of deviation, the original predicted additional bag retrieval quantity is dynamically adjusted. Specifically, if the actual inventory is lower than the theoretical inventory, indicating that actual consumption or outbound shipments are more than expected, the predicted additional bag retrieval quantity is increased to make up for the inventory gap. Conversely, if the actual inventory is higher than the theoretical inventory, the predicted additional bag retrieval quantity is reduced to avoid overstocking. The adjusted predicted additional bag retrieval task quantity will cover the original forecast value, forming a new forecast plan. Subsequently, the edge computing unit synchronously sends the updated predicted additional bag retrieval task data to the bag material warehouse system and the cloud central scheduling platform, ensuring that upstream inventory allocation and cross-node transfer calculations can make optimized decisions based on the latest inventory status and demand forecasts, thereby improving the responsiveness and inventory management accuracy of the entire supply chain.

[0099] In this embodiment, the present invention provides an automatic bag-retrieving machine inventory real-time synchronization system, comprising: The prediction module is used to extract HIS task data and historical bag retrieval data, generate internal bag retrieval tasks and predict additional bag retrieval tasks for the next bag retrieval application cycle, summarize them to obtain the bag types and quantities for the next bag retrieval application cycle, and then send them to the bag material warehouse system to apply for inventory for the next bag retrieval application cycle.

[0100] The matching module is used in the bag retrieval application cycle. When the bag retrieval task is executed, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the incorrect bag retrieval is blocked. When the verification result is a match, the bag retrieval is executed, and the bag retrieval event data is generated and input into the processing engine.

[0101] The compression module is used to obtain the number of outbound events within a preset time window, merge high-frequency continuous outbound events into batch events, and perform data compression and index storage.

[0102] The update module is used to obtain the measured inventory based on RFID and sensor information, and the theoretical inventory based on the bag retrieval task record. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds the threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag retrieval task for the next bag retrieval application cycle.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0104] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementation methods. Clearly, many modifications and variations can be made based on the content of this specification. The selection and detailed description of these embodiments in this specification are intended to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. Any modifications or variations that do not deviate from the structure of the invention or exceed the scope defined by the invention should fall within the protection scope of the invention.

Claims

1. A method for real-time inventory synchronization of an automatic bag-retrieving machine, characterized in that, include: The automatic bag-picking machine has a built-in edge computing unit that extracts HIS task data and historical bag-picking data, generates internal bag-picking tasks and predicted additional bag-picking tasks for the next bag-picking application cycle, summarizes them to obtain the bag types and quantities for the next bag-picking application cycle, and then sends them to the bag material warehouse system to apply for inventory for the next bag-picking application cycle. During the bag retrieval application cycle, when the bag retrieval task is executed, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the incorrect bag retrieval is blocked. When the verification result is a match, the bag retrieval is executed, and the bag retrieval event data is generated and input into the processing engine. Get the number of outbound events within a preset time window, merge high-frequency consecutive outbound events into batch events, and perform data compression and index storage; The measured inventory is obtained based on RFID and sensor information, and the theoretical inventory is obtained based on bag retrieval task records. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds the threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag retrieval tasks for the next bag retrieval application cycle.

2. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The specific process for generating the internal bag-retrieving task and predicting additional bag-retrieving tasks for the next bag-retrieving application cycle is as follows: The automatic bag-retrieving machine has a built-in edge computing unit that extracts HIS task data for the next bag-retrieving application cycle from the HIS. The HIS task data includes the bag type code, expected bag quantity, and expected bag-retrieving time of the hospital area where the automatic bag-retrieving machine is located. This data is used to obtain the internal bag-retrieving task for the next bag-retrieving application cycle. The internal bag-retrieving task includes the bag type code, bag quantity, and expected outbound channel number. These data are concatenated and combined according to a preset field order and data format to form a target triplet hash value. The automatic bag-retrieving machine has a built-in edge computing unit that extracts historical bag-retrieving data, including the number of historical inventory drift events, the number of historical emergency dispatches, and the number of historical inventory shortages. Based on the historical bag-retrieving data, historical bag-retrieving feature values ​​are generated, and based on the historical bag-retrieving feature values, the predicted additional bag-retrieving tasks for the next bag-retrieving application cycle are predicted. The system summarizes the internal bag-retrieving tasks and predicted additional bag-retrieving tasks for the next bag-retrieving application cycle of the automatic bag-retrieving machine to obtain the bag types and quantities for the next bag-retrieving application cycle, and then sends the information to the bag material warehouse system to apply for inventory for the next bag-retrieving application cycle.

3. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The edge computing unit performs matching and verification of the bag retrieval task, specifically as follows: If it is an internal bag retrieval task, during the outbound execution phase, the built-in RFID reading module of the automatic bag retrieval machine uses high-frequency radio frequency identification to perform non-contact scanning on the bag box located at the outbound channel waiting position, collect the bag type code of the bag box in real time, and transmit the bag type code to the edge computing unit. The edge computing unit extracts the bag type code from the currently executing bag picking task, and at the same time obtains the channel number of the currently executing outbound operation; The edge computing unit concatenates and combines the bag type code, bag quantity, and outbound channel number of the bag box according to the preset field order and data format to form triplet index data, and caches the triplet data in a temporary verification queue. The edge computing unit performs hash value calculation on the triplet index data and matches and verifies it with the target triplet hash value pre-generated and stored by the internal bag retrieval task. If the hash matching result does not match, the edge computing unit immediately sends a blocking command to the outbound execution module to control the outbound drive mechanism to stop its operation and trigger error event logging. If the hash match is successful, the edge computing unit sends an execution instruction to the outbound execution module to drive the corresponding channel to complete the material outbound process. After the outbound process is completed, internal outbound event data containing task ID, bag type code, outbound channel number and outbound time is generated and fed into the processing engine in a streaming manner.

4. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The edge computing unit performs matching and verification of the bag-retrieving task, and also includes: If it is an additional bag retrieval task, the edge computing unit calculates the available predicted reserve of the bag material type in the current bag retrieval application cycle and matches it with the target bag retrieval quantity of the additional bag retrieval task. If the available predicted reserve is insufficient, an inventory shortage reminder is sent to the automatic bag retrieval machine display terminal, and the data of the additional bag retrieval task is uploaded to the cloud central dispatch platform to generate an emergency dispatch event. If there is sufficient available forecast margin, the material will be discharged from the warehouse, driving the corresponding discharge channel to complete the discharge of the bagged material, generating additional discharge event data, and inputting the additional discharge event data into the processing engine in a streaming manner.

5. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The process of obtaining the number of outbound events within a preset time window, merging high-frequency consecutive outbound events into batch events, and performing data compression and index storage specifically includes: The edge computing unit continuously monitors the arrival timestamps of events from the streaming data stream of outbound events, based on a sliding time window mechanism, and counts the number of outbound events for various types of bag codes within a preset time window; When the number of times a certain type of bagged material leaves the warehouse exceeds a preset high-frequency threshold within a continuous time window, adjacent events within that time window are merged to generate a single batch event record. The batch event records the bagged material type code, batch quantity, corresponding task ID sequence, and set of warehouse exit channel numbers. After generating batch events, the processing engine performs structured compression on the event data, replacing repetitive fixed fields with a single-time storage method, and saving dynamic fields in differential encoding form.

6. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The process of obtaining measured inventory based on RFID and sensor information, and theoretical inventory based on bag-retrieving task records, specifically includes: During each bag-picking application cycle, the edge computing unit collects in real time the unique identifier of the bag box identified by the built-in RFID module of the automatic bag picking machine and the occupancy status of the bag box in the warehouse detected by the sensor, and calculates the actual inventory quantity of various types of bags. Meanwhile, based on the bag retrieval task records within the bag retrieval application cycle, which include the execution records of internal bag retrieval tasks and predicted additional bag retrieval tasks, the theoretical inventory quantity of the corresponding bag material type code is calculated by accumulating the number of tasks.

7. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 1, characterized in that: The edge computing unit compares the measured inventory with the theoretical inventory, specifically including: The edge computing unit performs a difference comparison between the measured inventory quantity and the theoretical inventory quantity, calculates the inventory deviation value, and if the absolute value of the inventory deviation value exceeds the preset inventory deviation threshold, an inventory drift event is immediately generated. The inventory drift event includes the bag type code, measured inventory quantity, theoretical inventory quantity, inventory deviation value, trigger time, and trigger reason code. The inventory drift event is input into the edge computing unit to update the predicted additional bag retrieval task for the next bag retrieval application cycle. Specifically, the original predicted additional bag retrieval quantity is corrected according to the deviation direction. The correction result is used to update the predicted additional bag retrieval task for the next cycle and synchronized to the bag material warehouse system and the cloud central scheduling platform for inventory allocation and transfer calculation.

8. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 4, characterized in that: The specific processing conditions for generating the emergency dispatch event are as follows: After an emergency dispatch event is generated, the edge computing unit uploads it to the central dispatch platform in the cloud. After receiving an emergency dispatch event, the cloud-based central dispatch platform obtains real-time inventory snapshots of other automatic bag-retrieving machines and information on the remaining bag-retrieving tasks within the current bag-retrieving application cycle, and calculates the available capacity of each other automatic bag-retrieving machine. After the adjustable reserve is calculated, the cloud-based central dispatch platform optimizes the allocation by taking into account the geographical distance between other automatic bag-retrieving machines and event-triggered devices, the allocation time, and the sufficiency of the reserve, generates temporary allocation tasks, and sends them to the support automatic bag-retrieving machines for execution.

9. The method for real-time inventory synchronization of an automatic bag-retrieving machine according to claim 8, characterized in that: The generation of emergency scheduling events also includes: if the total output of a certain type of bag material from the automatic bag-retrieving machine exceeds a sudden threshold, the edge computing unit generates an emergency scheduling event, with the specific processing conditions being: During the execution of the bag retrieval application cycle, when the cumulative outbound quantity of a certain bag material type code exceeds the preset sudden threshold within the preset bag retrieval time window, the edge computing unit immediately generates an emergency scheduling event. After an emergency dispatch event is generated, the edge computing unit immediately performs local processing, pushes the emergency bag usage warning information to the automatic bag dispensing machine display terminal, and uploads it to the cloud central dispatch platform for allocation and optimization processing; Simultaneously, extract the remaining internal bag retrieval tasks within the current bag retrieval application cycle, obtain the required bag types and quantities for the remaining internal bag retrieval tasks, compare them with the local inventory, and if the local inventory can meet the required bag types and quantities for the remaining internal bag retrieval tasks, obtain the inventory balance between the required bag types and quantities for the remaining internal bag retrieval tasks and the local inventory, and send the inventory balance to the cloud central scheduling platform to prioritize the allocation optimization. If the local inventory cannot meet the required bag types and quantities for the remaining internal bag retrieval tasks, obtain the inventory shortage between the required bag types and quantities for the remaining internal bag retrieval tasks and the local inventory, and send the inventory shortage to the cloud central scheduling platform to prioritize the allocation optimization.

10. A system applying the real-time inventory synchronization method for an automatic bag-retrieving machine as described in any one of claims 1-9, characterized in that: The prediction module is used to extract HIS task data and historical bag retrieval data, generate internal bag retrieval tasks and predict additional bag retrieval tasks for the next bag retrieval application cycle, summarize them to obtain the bag retrieval types and quantities for the next bag retrieval application cycle, and then send them to the bag material warehouse system to apply for inventory for the next bag retrieval application cycle. The matching module is used in the bag retrieval application cycle. When the bag retrieval task is executed in the warehouse, the edge computing unit performs matching verification on the bag retrieval task. When the verification result is a mismatch, the incorrect warehouse exit is blocked. When the verification result is a match, the warehouse exit is executed, and the warehouse exit event data is generated and input into the processing engine. The compression module is used to obtain the number of outbound events within a preset time window, merge high-frequency continuous outbound events into batch events, and perform data compression and index storage. The update module is used to obtain the measured inventory based on RFID and sensor information, and the theoretical inventory based on the bag retrieval task record. The edge computing unit compares the measured inventory with the theoretical inventory. If the inventory deviation exceeds the threshold, it is recorded as an inventory drift event and input into the edge computing unit to update the predicted additional bag retrieval task for the next bag retrieval application cycle.

Images