An unmanned medical consumable warehouse management system and method

Abstract

This invention relates to the field of unmanned warehouse management technology, and more particularly to an unmanned medical consumables warehouse management system and method. Specifically, it includes a warehouse management center, a data acquisition and verification module, a location allocation module, a location dynamic adjustment module, a periodic inventory module, and a backend early warning module. On one hand, it automatically collects consumables information and completes preliminary verification to ensure no data omissions. On the other hand, through multi-angle feature image acquisition and defect identification, it accurately identifies appearance problems of consumables. Furthermore, through multi-dimensional scoring and comprehensive ranking based on factors such as ease of access and category concentration, it selects the optimal location. Simultaneously, by accurately identifying consumables nearing their expiration date and high-frequency consumables requiring adjustment, it promptly adjusts consumables with insufficient access convenience or excessively long retrieval times to better locations. Finally, through the autonomous movement of inventory robots and the fusion positioning technology of LiDAR and visual navigation, it achieves unmanned automatic inventory management of all warehouse locations, avoiding stockouts or overstocking.

Description

Technical Field

[0001] This invention relates to the field of unmanned warehouse management technology, and in particular to an unmanned medical consumables warehouse management system and method. Background Technology

[0002] Against the backdrop of rapid development in the medical industry, medical consumables, as core supporting materials for clinical diagnosis and treatment, surgical procedures, etc., are experiencing continuous growth in demand and increasingly complex categories (covering multiple categories such as disposable medical devices, reagents, and excipients). This has led to ever-increasing requirements for the refinement, efficiency, and precision of warehouse management. The quality of medical consumable management is directly related to the safety and continuity of clinical diagnosis and treatment, as well as the control of hospital operating costs and the efficiency of resource utilization. It is a key link in the hospital's logistics support system.

[0003] However, the current management of medical consumables warehouses still faces several pain points:

[0004] Lack of intelligent planning in warehouse location allocation: Warehouse location allocation relies heavily on the experience of management personnel, resulting in low accuracy and difficulty in optimizing the allocation of warehouse resources. This further reduces the efficiency of inbound and outbound operations. For example, consumables with only 2 months of remaining shelf life are allocated to warehouse locations that are difficult to access, causing slow turnover and eventually being scrapped due to expiration.

[0005] The periodic inventory process is cumbersome and prone to errors: Traditional inventory methods require a lot of manpower to manually check each location and each category, resulting in long inventory cycles and high error rates. Furthermore, the comparison between inventory results and inventory data relies heavily on manual processing, making it difficult to quickly identify issues such as discrepancies in categories, quantities, expiration dates, and misplaced locations. This makes it impossible to provide timely and accurate data support for inventory optimization and consumable scheduling.

[0006] Therefore, developing an unmanned medical consumables warehouse management system and method to solve the pain points of low efficiency, large errors and poor coordination in the traditional management model, and to achieve refined and intelligent management of medical consumables warehouses, has become an urgent need in the current medical logistics support field. Summary of the Invention

[0007] The purpose of this invention is to provide an unmanned medical consumables warehouse management system and method. This system accurately identifies appearance problems with consumables through data collection and verification, simultaneously verifies the consistency between the quantity of consumables and the statistics of electronic tags in real time, and selects the optimal storage location through multi-dimensional scoring and comprehensive ranking. Furthermore, it accurately identifies consumables nearing their expiration date and high-frequency consumables that require adjustment, promptly moving consumables with insufficient access convenience or excessively long retrieval times to better storage locations. Finally, through the autonomous movement of inventory robots and the fusion positioning technology of LiDAR and visual navigation, it achieves unmanned automatic inventory management of all warehouse locations, avoiding stockouts or overstocking.

[0008] The objective of this invention can be achieved through the following technical solution: an unmanned medical consumables warehouse management system, comprising a warehouse management center, a data acquisition and verification module, a storage location allocation module, a storage location dynamic adjustment module, a periodic inventory module, and a backend early warning module;

[0009] The data acquisition and verification module is used to perform preliminary verification on the electronic tag information of the consumable packaging and the multi-angle feature images of the consumables after the consumables are placed, and output deviation instructions or inventory file numbers.

[0010] The storage location allocation module is used to extract the inventory file number to obtain the consumable attribute information, filter the candidate storage location set from the available storage locations based on the preset consumable allocation rules, calculate and sort the comprehensive score of the multi-dimensional scoring index of the candidate storage locations, select the optimal storage location, construct the storage location-consumable association file and send it to the warehouse management center for storage.

[0011] The dynamic storage location adjustment module is used to mark occupied storage locations based on the storage location-consumables association file, collect complete data files of consumables in occupied storage locations and real-time warehouse status data to form a basic dataset for investigation, filter consumables that need to be adjusted for near-expiry dates and high-frequency consumables that need to be adjusted, construct a storage location adjustment list, and trigger the backend early warning module to display the list.

[0012] The periodic inventory module is used to respond to the preset inventory cycle. After checking that all equipment is ready, it generates an inventory instruction, synchronizes the latest inventory data as the inventory baseline data, controls the inventory robot to move along the preset optimal path and corrects the trajectory, reads electronic tags and collects images of consumables in the storage location, generates comprehensive actual data to compare with the inventory baseline data, records an anomaly feedback list and completes the full storage location inventory.

[0013] Preferably, the analysis process of the data acquisition and verification module is as follows:

[0014] The existing RFID reader automatically reads the electronic tag information on the consumable packaging to obtain the core identity data of the consumables. At the same time, it performs preliminary verification of the electronic tag information. If the tag reading fails, the data is incomplete, or the reading is repeated, a reading failure command is triggered and an audible and visual prompt is issued immediately until the information of all consumables is successfully collected.

[0015] The existing high-definition vision sensor is used to acquire multi-angle feature images of consumables in the recognition area. The acquired multi-angle feature images are preprocessed and then input into a preset image recognition model to output defect recognition results (defect present / no defect). At the same time, the number of individual consumables in the image is counted, and the defect recognition results and the number of individual consumables are analyzed to output deviation instructions or valid instructions.

[0016] Preferably, when a valid instruction is generated, the identity core data, multi-angle feature images, and individual quantities of consumables are associated and bound together with a unique identifier. A preset inventory file template is then called, and the associated information is filled into the preset inventory file template to generate a unique inventory file.

[0017] Preferably, the analysis process of the cargo location allocation module is as follows:

[0018] Extract the inventory file number of the consumables to be allocated in the current storage location, obtain the attribute information of the consumables to be allocated in the current storage location based on the inventory file number, obtain the preset consumable allocation rules, including priority rules and adaptability rules, and based on the adaptability rules, filter out the candidate storage location set that meets the current consumable category, environmental requirements, size and weight restrictions from all available storage locations.

[0019] Preferably, a multi-dimensional scoring index is obtained for each storage location in the candidate storage location set. The multi-dimensional scoring index includes a score for ease of access and a score for category concentration.

[0020] Obtain the preset weight coefficients of each indicator in the multidimensional scoring index, and calculate the comprehensive score = ∑ (each indicator × corresponding preset weight coefficient).

[0021] The candidate storage locations are sorted from highest to lowest based on their overall scores. The storage location with the highest overall score is selected as the optimal storage location. If there is not a unique number of storage locations with the highest overall scores, the storage location with the shortest robot retrieval path time is selected as the optimal storage location.

[0022] Preferably, the analysis process of the dynamic adjustment module for cargo space is as follows:

[0023] Based on the location-consumables association file, the corresponding location of the consumables is marked as an occupied location, the complete data file of the consumables in the occupied location is obtained, and the real-time status data of the warehouse is collected simultaneously to form a basic dataset for investigation.

[0024] Filter out all near-expiration consumables in occupied storage locations, extract the accessibility score of the currently occupied storage location for this type of consumable, analyze the accessibility score, and mark the consumables in occupied storage locations with accessibility scores lower than the preset accessibility score threshold as near-expiration consumables that need to be adjusted.

[0025] Preferably, the average daily usage frequency of each consumable over the past 30 days is calculated and compared with the average daily usage frequency over the historical 90 days. Consumables with a frequency increase greater than or equal to a preset threshold are selected. The accessibility score of the current storage location and the time taken for the robot to retrieve the goods are checked. If the accessibility score is less than the preset accessibility score threshold, or the time taken for the retrieval path is greater than the preset time, then the consumable is determined to be a high-frequency consumable that needs to be adjusted.

[0026] A storage location adjustment list is constructed based on the need to adjust near-expiry consumables, the need to adjust high-frequency consumables, and the storage location-consumable association file.

[0027] Preferably, the analysis process of the periodic inventory module is as follows:

[0028] S1: Obtain the real-time status of each inspection device. The real-time status includes ready and pending. If the real-time status of each inspection device is ready, then generate an inventory command.

[0029] S2: Responds to inventory count instructions by synchronizing the latest current inventory data as the baseline data for inventory count;

[0030] S3: The inventory robot moves autonomously along the preset optimal path. When it deviates from the preset path by less than or equal to the preset path threshold, it triggers a control command and automatically fine-tunes its direction to return to the path.

[0031] S4: After the robot arrives at the target storage location, it adjusts its position according to the preset posture angle and reads the electronic tags of the consumables in the storage location and collects multi-angle feature images.

[0032] S5: Analyze the electronic tag information and multi-angle feature images of consumables in the storage location, generate a verification result list, and integrate it with the electronic tag information collected by RFID to form comprehensive actual data of the storage location;

[0033] S6: Compare the actual data with the inventory baseline data one by one, including comparison of type, quantity, expiration date, packaging and location. If they are all consistent, it is judged as normal and a continue instruction is triggered. If they are not all consistent, it is judged as abnormal and the inconsistencies and storage location information are recorded to form a feedback list.

[0034] S7: In response to the continue command, automatically move to the next target location for inventory counting, and mark the inventory counted location as an inventory counted location, until all target locations have been inventoried.

[0035] The present invention also proposes an unmanned medical consumables warehouse management method, comprising the following steps:

[0036] Step 1: Data collection and verification: After the consumables are placed, the collected electronic tag information is initially verified until all consumable information is successfully collected. Then, the multi-angle feature images of the collected consumables are identified and analyzed, and deviation instructions or inventory file numbers are output.

[0037] Step 2: Location Allocation: Based on the preset consumable allocation rules, a set of candidate locations is selected from all available locations, and the optimal location is selected by calculating the comprehensive score of the candidate locations;

[0038] Step 3: Dynamic Adjustment of Storage Locations: Construct a basic dataset for investigating occupied storage locations, filter out consumables nearing their expiration date and high-frequency consumables that need adjustment, construct a storage location adjustment list, and trigger an alert display;

[0039] Step 4: Periodic Inventory: Based on the preset inventory cycle, conduct a preliminary check on the actual status of each device until all devices are ready, then generate an inventory command to conduct a consumable inventory.

[0040] Step 5: Correction and Comparison: Real-time trajectory correction is performed when the inventory robot moves along the preset optimal path. Electronic tags are read and images are collected from consumables in the storage location. Comprehensive actual data is generated and compared with the inventory baseline data. An anomaly feedback list is recorded and the full storage location inventory is completed.

[0041] The beneficial effects of this invention are as follows:

[0042] (1) On the one hand, it automatically collects consumable information and completes preliminary verification to ensure that no data is missed. On the other hand, it accurately investigates appearance problems such as damage and deformation of consumables through multi-angle feature image collection and defect identification. At the same time, it verifies the consistency between the quantity of consumables and the statistical quantity of electronic tags. It ensures the authenticity and good condition of the consumable information entering the warehouse from both data and physical dimensions, laying a reliable foundation for subsequent management. Furthermore, it selects the optimal storage location through multi-dimensional scoring and comprehensive ranking based on storage and retrieval convenience, category concentration, etc. It also supports secondary screening of the path time of the optimal scoring storage location, reducing the cost of robot storage and retrieval path and improving the efficiency of inbound and outbound operations.

[0043] (2) By accurately identifying the near-expiry consumables and high-frequency consumables that need adjustment, consumables with insufficient access or excessively long retrieval time can be promptly adjusted to better storage locations. This ensures the rapid turnover of near-expiry consumables and reduces expired losses, while also meeting the high-efficiency storage and retrieval needs of high-frequency consumables, reducing the cost of manual handling and robot operation. Furthermore, through the autonomous movement of the inventory robot and the fusion positioning technology of LiDAR and visual navigation, the warehouse can achieve unmanned automatic inventory of all storage locations, which solves the problems of low efficiency, large errors, and high labor intensity of manual inventory, and avoids stockouts or backlogs. Attached Figure Description

[0044] The invention will now be further described with reference to the accompanying drawings;

[0045] Figure 1 This is a flowchart of the system of the present invention;

[0046] Figure 2 This is a reference diagram of the method of the present invention;

[0047] Figure 3 This is a reference diagram for analyzing the cargo location adjustment list 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 this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments;

[0050] Example 1: Please refer to Figures 1 to 3 As shown, this invention is an unmanned medical consumables warehouse management system, including a warehouse management center, a data acquisition and verification module, a storage location allocation module, a storage location dynamic adjustment module, a periodic inventory module, and a backend early warning module. The warehouse management center has bidirectional communication connections with the data acquisition and verification module and the storage location allocation module, and a unidirectional communication connection with the storage location dynamic adjustment module. The data acquisition and verification module and the storage location dynamic adjustment module both have unidirectional communication connections with the backend early warning module, and the warehouse management center has a unidirectional communication connection with the periodic inventory module.

[0051] The data acquisition and verification module is used to perform preliminary verification of the electronic tag information of the consumable packaging and the multi-angle feature images of the consumables after placement, and outputs deviation instructions or inventory file numbers, specifically including:

[0052] By using traditional RFID readers to automatically read the electronic tag information on the consumable packaging, core identification data such as consumable name, specifications, manufacturer, expiration date, and batch number can be obtained.

[0053] At the same time, the electronic tag information is initially verified. If the tag reading fails, the data is incomplete, or the reading is repeated, a reading failure command is triggered, and an audible and visual prompt is immediately issued. The staff then rearranges the consumables or checks the tag status (such as the tag falling off or being damaged) until the information of all consumables is successfully collected.

[0054] The existing high-definition vision sensor is used to collect multi-angle feature images of consumables within the recognition area. The collection angle covers the front, side and top of the consumables. The number of images collected for each batch of consumables is ≥10 frames to ensure no visual blind spots.

[0055] The acquired multi-angle feature images are preprocessed (e.g., noise reduction, enhancement), and then input into a preset image recognition model. The model outputs a defect recognition result (defect present / no defect). Simultaneously, the number of individual consumables in the image is counted, and the defect recognition result and the number of individual consumables are analyzed. If the defect recognition result is defective (including damage, deformation, stains, broken seals, etc.), or if the number of individual consumables is not equal to the electronic tag information statistics (representing the total number of times the RFID reader reads the electronic tag information on the consumable packaging), a deviation command is generated. If the defect recognition result is no defect, and the number of individual consumables is equal to the electronic tag information statistics, a valid command is generated. The backend early warning module responds to the deviation command and immediately displays the preset early warning text corresponding to the deviation command, so as to manage the consumables in a timely manner and ensure the validity and completeness of the information of the consumables in the current allocation location.

[0056] When a valid instruction is generated, the core identity data, multi-angle feature images, and individual quantities of consumables are associated and bound together, and a unique identifier is assigned. A preset inventory file template is called, and the associated information is filled into the preset inventory file template to generate a unique inventory file containing basic consumable information, warehousing time, supplier information, storage location reservation (temporary storage status), appearance inspection results, quantity, expiration date warning threshold, etc. The inventory file number is generated by year + month + date + batch number + serial number, and the inventory file number is sent to the warehouse management center for storage.

[0057] The location allocation module is used to extract inventory file numbers to obtain consumable attribute information, filter candidate locations from available locations based on preset consumable allocation rules, calculate and sort the comprehensive scores of the candidate locations using multi-dimensional scoring indicators, select the optimal location, construct a location-consumable association file, and send it to the warehouse management center for storage. Specifically, this includes:

[0058] Extract the inventory file number of the consumables in the current location to be allocated, and obtain the attribute information of the consumables in the current location to be allocated based on the inventory file number. The attribute information includes basic information of the consumables, the time of entry into the warehouse, etc.

[0059] Retrieve preset consumable allocation rules, including priority rules and compatibility rules.

[0060] The priority rules are as follows: Consumables nearing their expiration date (remaining validity period ≤ 3 months) priority > High-frequency used consumables (daily average usage ≥ 5 times) priority > Consumables in urgent need priority > Ordinary consumables priority;

[0061] Adaptability rules: Consumable environmental requirements must be fully matched with the storage location's environmental capabilities (e.g., low-temperature consumables are only compatible with refrigerated storage locations), packaging size ≤ available storage space, weight ≤ storage location's weight-bearing limits, etc.

[0062] Based on the adaptability rules, a set of candidate storage locations that meet the current consumable category, environmental requirements, and size and weight restrictions is selected from all available storage locations.

[0063] Obtain multi-dimensional scoring indicators for each storage location in the candidate storage location set. These multi-dimensional scoring indicators include accessibility scores, category concentration scores, etc.

[0064] For example: Accessibility is represented by the sum of the distances between the storage location and the outbound aisle, and the robot's starting point. The closer the distance, the better. Accessibility score = 100 - [(distance from storage location to outbound aisle × 0.6 + distance from storage location to robot's starting point × 0.4) / D0 × 100]. If the calculated result is negative, it is scored as 0 points; if it exceeds 100 points, it is scored as 100 points. The benchmark distance D0 is the average distance from the storage location in the warehouse to the outbound aisle (e.g., 5 meters), which is used as the scoring benchmark.

[0065] Category concentration score = 100 × (number of storage locations occupied by the same category Sd / total number of storage locations in the area S). Note: If there are no consumables of the same category in the target area (Sd = 0), the score is 0 points; if all storage locations in the target area store consumables of the same category (Sd = S), the score is 100 points.

[0066] Example: The total number of storage locations in the candidate storage location area is S = 10, of which 7 storage locations of the same type of consumables are already occupied by Sd = 7; Score = 100 × (7 / 10) = 70 points; If Sd = 0, the score = 0 points;

[0067] Obtain the preset weight coefficients of each indicator in the multidimensional scoring index, and calculate the comprehensive score = ∑ (each indicator × corresponding preset weight coefficient).

[0068] The candidate storage locations are sorted from highest to lowest based on their overall scores, and the storage location with the highest overall score is selected as the optimal storage location.

[0069] If the number of highest comprehensive scores is not unique, the storage location with the shortest robot retrieval path time will be taken as the optimal storage location. Based on the optimal storage location and the consumables, a storage location-consumables association file will be constructed and sent to the warehouse management center for storage, so that the unmanned robot can accurately place the consumables.

[0070] Example 2: The dynamic location adjustment module is used to mark occupied locations based on the location-consumable association file, collect complete data files of consumables in occupied locations and real-time warehouse status data to form a basic dataset for investigation, filter consumables nearing their expiration date and high-frequency consumables that need adjustment, construct a location adjustment list, and trigger the backend early warning module to display the list, specifically including:

[0071] Based on the location-consumable association file, the location corresponding to the consumable is marked as an occupied location, and the complete data file of the consumable in the occupied location is obtained, including the basic attributes of the consumable (such as category, specifications, value), dynamic attributes (such as remaining shelf life, usage frequency in the past 30 days, historical access records), and current storage information (such as location number, location area, original allocation compatibility score), etc.

[0072] Synchronously collect real-time warehouse status data: storage location occupancy rate (total number of storage locations in the area ÷ number of storage locations occupied × 100%), accessibility scores for each storage location, etc., to form a basic dataset for investigation;

[0073] Filter out all near-expiration consumables in occupied storage locations (meaning remaining validity period ≤ 3 months), extract the accessibility score of the currently occupied storage location for this type of consumable, analyze the accessibility score, mark the consumables in occupied storage locations with accessibility scores lower than the preset accessibility score threshold as near-expiration consumables that need adjustment, and record the original storage location number, remaining validity period and current accessibility score to form an adjustment data package;

[0074] Calculate the average daily usage frequency of each consumable over the past 30 days and compare it with the average daily usage frequency over the historical 90 days to select consumables whose frequency increase is greater than or equal to a preset threshold.

[0075] Check the accessibility score of the current storage location and the time taken for the robot to pick up the goods. If the accessibility score is less than the preset accessibility score threshold, or the time taken for the picking up the goods is greater than the preset time, then the consumable is determined to be a high-frequency consumable that needs to be adjusted.

[0076] Based on the need to adjust near-expiry consumables, the need to adjust high-frequency consumables, and the location-consumable association files, a location adjustment list is constructed. The backend early warning module is used to respond to the location adjustment list and immediately display the location adjustment list so that operation and management personnel can make regular adjustments to consumables based on the location adjustment list, thereby improving the management efficiency of consumables.

[0077] The periodic inventory module responds to preset inventory cycles. After checking that all equipment is ready, it generates an inventory command, synchronizes the latest inventory data as the inventory baseline data, controls the inventory robot to move along a preset optimal path and corrects its trajectory, reads electronic tags and collects images of consumables in the storage locations, generates comprehensive actual data for comparison with the inventory baseline data, records an anomaly feedback list, and completes a full-location inventory count, as detailed below:

[0078] S1: Obtain the real-time status of each inspection device (including RFID reader, vision sensor, etc.). The real-time status includes ready and pending. If the real-time status of each inspection device is ready, generate an inventory instruction.

[0079] S2: Respond to inventory count instructions to synchronize the latest inventory data as the basis for inventory count: including the recorded quantity, storage location, and expiration date information of each consumable;

[0080] S3: The inventory robot moves autonomously along the preset optimal path and corrects its trajectory in real time by combining LiDAR and visual navigation: when it deviates from the preset path ≤ the preset path threshold, it triggers an adjustment command and automatically fine-tunes the direction to return to the path.

[0081] S4: After the robot arrives at the target storage location, it adjusts its position according to the preset posture angle and reads the electronic tags of the consumables in the storage location and collects multi-angle feature images.

[0082] S5: Analyze the electronic tag information and multi-angle feature images of consumables in the storage location, generate a verification result list, which includes the actual quantity, location status, packaging status, etc., and integrates it with the electronic tag information collected by RFID to form the comprehensive actual data of the storage location.

[0083] S6: Compare the actual data with the inventory baseline data one by one, including comparison of type, quantity, expiration date, packaging and location. If they are all consistent, it is judged as normal and a continue instruction is triggered. If they are not consistent, it is judged as abnormal and the inconsistencies, storage locations and other information are recorded to form a feedback list.

[0084] Category comparison: Verify whether the actual types of consumables are consistent with the latest inventory data;

[0085] Quantity comparison: Compare the actual quantity of each consumable with the quantity recorded in the latest inventory data;

[0086] Expiry date comparison: Verify whether the actual expiry date is consistent with the latest inventory data record, and determine whether there are consumables nearing their expiry date (indicating remaining expiry date ≤ 3 months) or expired consumables;

[0087] Location comparison: Confirm whether the actual storage location is consistent with the location of the latest inventory data record;

[0088] Packaging comparison: Record any abnormalities such as damaged or opened packaging;

[0089] S7: In response to the continue command, automatically move to the next target location for inventory counting, and mark the inventory counted location as an inventory counted location, until all target locations have been inventoried.

[0090] Example 3: This invention also proposes an unmanned medical consumables warehouse management method, including the following steps:

[0091] Step 1: Data collection and verification: After the consumables are placed, the collected electronic tag information is initially verified until all consumable information is successfully collected. Then, the multi-angle feature images of the collected consumables are identified and analyzed, and deviation instructions or inventory file numbers are output.

[0092] Step 2: Location Allocation: Based on the preset consumable allocation rules, a set of candidate locations is selected from all available locations, and the optimal location is selected by calculating the comprehensive score of the candidate locations;

[0093] Step 3: Dynamic Adjustment of Storage Locations: Construct a basic dataset for investigating occupied storage locations, filter out consumables nearing their expiration date and high-frequency consumables that need adjustment, construct a storage location adjustment list, and trigger an alert display;

[0094] Step 4: Periodic Inventory: Based on the preset inventory cycle, conduct a preliminary check on the actual status of each device until all devices are ready, then generate an inventory command to conduct a consumable inventory.

[0095] Step 5: Correction and Comparison: Real-time trajectory correction is performed when the inventory robot moves along the preset optimal path. Electronic tags are read and images are collected from consumables in the storage location. Comprehensive actual data is generated and compared with the inventory baseline data. An anomaly feedback list is recorded and the full storage location inventory is completed.

[0096] In summary, on the one hand, the system automatically collects consumable information and performs preliminary verification to ensure no data omissions. On the other hand, through multi-angle feature image acquisition and defect identification, it accurately identifies appearance problems such as damage and deformation of consumables. Simultaneously, it verifies the consistency between the quantity of consumables and the statistical data on electronic tags. This dual approach, combining data and physical inventory, ensures the authenticity and good condition of incoming consumable information, laying a reliable foundation for subsequent management. Furthermore, through multi-dimensional scoring and comprehensive ranking based on factors such as ease of access and category concentration, the system selects the optimal storage location. It also supports secondary filtering of the path time for the optimal-scoring storage location, reducing the robot's access path. This reduces costs and improves the efficiency of inbound and outbound operations. By accurately identifying consumables nearing their expiration date and high-frequency consumables that need adjustment, consumables with insufficient access convenience or excessively long retrieval routes are promptly moved to better storage locations. This ensures the rapid turnover of consumables nearing their expiration date, reducing expired losses, while also meeting the high-efficiency storage and retrieval needs of high-frequency consumables. It also reduces the costs of manual handling and robot operations. Furthermore, through the autonomous movement of inventory robots and the fusion positioning technology of LiDAR and visual navigation, unmanned automated inventory counting is achieved in all warehouse locations. This solves the problems of low efficiency, large errors, and high labor intensity of manual inventory counting, and avoids stockouts or backlogs.

[0097] The threshold is set for comparative analysis of results to determine whether they are good or bad. The value of the threshold is determined by a combination of large-scale model analysis of sample data and human experience. It can also be adjusted appropriately based on seasonal or common-sense influencing factors.

[0098] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An unmanned medical consumable store management system, characterized by, The warehouse management center, the data acquisition and verification module, the goods location allocation module, the goods location dynamic adjustment module, the periodic inventory module and the backend early warning module are included. The data acquisition and verification module is used for completing preliminary verification of the collected consumable packaging electronic tag information and consumable multi-angle feature images after the consumables are placed, and outputting deviation instructions or warehouse record numbers. The goods location allocation module is used for extracting the warehouse record number to obtain consumable attribute information, screening a candidate goods location set from idle goods locations based on a preset consumable allocation rule, calculating a comprehensive score of candidate goods location multi-dimensional score indicators and sorting, selecting an optimal goods location, and constructing a goods location-consumable association file and sending it to the warehouse management center for storage. The goods location dynamic adjustment module is used for marking occupied goods locations based on the goods location-consumable association file, collecting consumable complete data files and warehouse real-time state data of the occupied goods locations to form an investigation basic data set, screening near-expiration consumables and high-frequency consumables that need to be adjusted, constructing a goods location adjustment list, and triggering the backend early warning module to display the list. The periodic inventory module is used for generating an inventory instruction after checking that all devices are ready in response to a preset inventory period, synchronizing the latest inventory data as inventory benchmark data, controlling the inventory robot to move along a preset optimal path and correct the trajectory, reading the electronic tags of the consumables in the goods locations and collecting images, comparing the comprehensive actual data with the inventory benchmark data, recording an abnormal feedback list, and completing the inventory of all goods locations.

2. The unmanned medical consumable store management system according to claim 1, characterized in that, The analysis process of the data acquisition and verification module is as follows: The existing RFID reader automatically reads the electronic tag information on the consumable packaging to obtain the identity core data of the consumables, and preliminarily verifies the electronic tag information. If the tag reading fails, the data is incomplete, or the reading is repeated, a reading failure instruction is triggered, and an audible and visual prompt is immediately issued until the information of all consumables is successfully collected. The existing high-definition visual sensor collects multi-angle feature images of the consumables in the identification area, pre-processes the collected multi-angle feature images, inputs the pre-processed multi-angle feature images into a preset image recognition model, outputs a defect recognition result (with / without defect), simultaneously counts the number of consumable individuals in the image, and analyzes the defect recognition result and the number of consumable individuals, and outputs a deviation instruction or a valid instruction.

3. The unmanned medical consumable store management system according to claim 2, wherein, When a valid instruction is generated, the identity core data, multi-angle feature images and consumable individual quantity are associated and bound with a unique identifier, a preset warehouse record template is called, the associated information after binding is filled into the preset warehouse record template, and a unique warehouse record is generated.

4. The unmanned medical consumable store management system according to claim 1, wherein, The analysis process of the goods location allocation module is as follows: The current goods location consumable warehouse record number is extracted, the attribute information of the current goods location consumable is obtained based on the warehouse record number, the preset consumable allocation rule is obtained, including the priority rule and the adaptability rule, and the candidate goods location set that meets the current consumable category, environmental demand, size and weight limit is screened from all idle goods locations based on the adaptability rule.

5. The unmanned medical consumable store management system according to claim 4, wherein, The multi-dimensional score indicators of each goods location in the candidate goods location set are obtained, including the access convenience score and the category concentration score. Obtain the preset weight coefficients of each indicator in the multidimensional scoring index, and calculate the comprehensive score = ∑ (each indicator × corresponding preset weight coefficient). The candidate storage locations are sorted from highest to lowest based on their overall scores. The storage location with the highest overall score is selected as the optimal storage location. If there is not a unique number of storage locations with the highest overall scores, the storage location with the shortest robot retrieval path time is selected as the optimal storage location.

6. The unmanned medical consumable store management system according to claim 1, wherein, The analysis process of the dynamic cargo location adjustment module is as follows: Based on the location-consumables association file, the corresponding location of the consumables is marked as an occupied location, the complete data file of the consumables in the occupied location is obtained, and the real-time status data of the warehouse is collected simultaneously to form a basic dataset for investigation. Filter out all near-expiration consumables in occupied storage locations, extract the accessibility score of the currently occupied storage location for this type of consumable, analyze the accessibility score, and mark the consumables in occupied storage locations with accessibility scores lower than the preset accessibility score threshold as near-expiration consumables that need to be adjusted.

7. The unmanned medical consumable store management system according to claim 1, wherein, Calculate the average daily usage frequency of each consumable over the past 30 days and compare it with the average daily usage frequency over the historical 90 days. Select consumables whose frequency increase is greater than or equal to a preset threshold. Check the accessibility score of the current storage location and the time taken for the robot to pick up the goods. If the accessibility score is less than the preset accessibility score threshold, or the time taken for the picking up the goods is greater than the preset time, then the consumable is determined to be a high-frequency consumable that needs to be adjusted. A storage location adjustment list is constructed based on the need to adjust near-expiry consumables, the need to adjust high-frequency consumables, and the storage location-consumable association file.

8. The unmanned medical consumable store management system according to claim 1, wherein, The analysis process of the periodic inventory module is as follows: S1: Obtain the real-time status of each inspection device. The real-time status includes ready and pending. If the real-time status of each inspection device is ready, then generate an inventory command. S2: Responds to inventory count instructions by synchronizing the latest current inventory data as the baseline data for inventory count; S3: The inventory robot moves autonomously along the preset optimal path. When it deviates from the preset path by less than or equal to the preset path threshold, it triggers a control command and automatically fine-tunes its direction to return to the path. S4: After the robot arrives at the target storage location, it adjusts its position according to the preset posture angle and reads the electronic tags of the consumables in the storage location and collects multi-angle feature images. S5: Analyze the electronic tag information and multi-angle feature images of consumables in the storage location, generate a verification result list, and integrate it with the electronic tag information collected by RFID to form comprehensive actual data of the storage location; S6: Compare the actual data with the inventory baseline data one by one, including comparison of type, quantity, expiration date, packaging and location. If they are all consistent, it is judged as normal and a continue instruction is triggered. If they are not all consistent, it is judged as abnormal and the inconsistencies and storage location information are recorded to form a feedback list. S7: In response to the continue command, automatically move to the next target location for inventory counting, and mark the inventory counted location as an inventory counted location, until all target locations have been inventoried.

9. An unmanned medical consumable warehouse management method, applied to the unmanned medical consumable warehouse management system of any one of claims 1-8, characterized in that, Includes the following steps: Step 1: Data collection and verification: After the consumables are placed, the collected electronic tag information is initially verified until all consumable information is successfully collected. Then, the multi-angle feature images of the collected consumables are identified and analyzed, and deviation instructions or inventory file numbers are output. Step 2: Location Allocation: Based on the preset consumable allocation rules, a set of candidate locations is selected from all available locations, and the optimal location is selected by calculating the comprehensive score of the candidate locations; Step 3: Dynamic Adjustment of Storage Locations: Construct a basic dataset for investigating occupied storage locations, filter out consumables nearing their expiration date and high-frequency consumables that need adjustment, construct a storage location adjustment list, and trigger an alert display; Step 4: Periodic Inventory: Based on the preset inventory cycle, conduct a preliminary check on the actual status of each device until all devices are ready, then generate an inventory command to conduct a consumable inventory. Step 5: Correction and Comparison: Real-time trajectory correction is performed when the inventory robot moves along the preset optimal path. Electronic tags are read and images are collected from consumables in the storage location. Comprehensive actual data is generated and compared with the inventory baseline data. An anomaly feedback list is recorded and the full storage location inventory is completed.

Images