An automatic stockyard management method and system based on real-time monitoring and alarm

Abstract

The application discloses an automatic material warehouse management method and system based on real-time monitoring and alarm, and relates to the technical field of medical oral material management. An industrial robot is controlled to grab raw materials to be stored from a feeding area, and the raw materials to be stored are detected to obtain raw material attribute information; a target position is determined in a plurality of positions of a material warehouse based on the raw material attribute information and a warehouse position allocation rule, the industrial robot is controlled to move the raw materials to be stored from the detection area to the target position, and the inventory is updated in real time; real-time inventory corresponding to a target raw material type is obtained from the material warehouse, historical demand data, procurement cycle data and average daily consumption corresponding to the target raw material type are obtained and calculated to obtain a dynamic safety inventory threshold; when the real-time inventory is less than or equal to the dynamic safety inventory threshold, replenishment warning information for the target raw material type is generated, and the replenishment warning information is sent to a terminal device of a person in charge in a communication mode. The technical scheme is implemented to improve the accuracy of oral material management.

Description

Technical Field

[0001] This application relates to the field of medical dental material management technology, specifically to an automated silo management method and system based on real-time monitoring and alarm. Background Technology

[0002] In medical device manufacturing, especially in the field of oral healthcare, the management of various materials, such as zirconia blocks and resins used in denture processing, or implants and orthodontic materials required for oral treatment, is directly related to the delivery cycle of customized medical products and the smooth progress of clinical treatment.

[0003] Existing automated management technologies for dental materials primarily utilize photoelectric sensors, RFID tags, or simple visual recognition to determine the presence of specific materials in the warehouse. These technologies are then integrated with warehouse management software to perform preliminary inventory checks. When the system detects that the inventory of a certain type of material is below a pre-set static threshold, it triggers an alarm, prompting management to replenish the stock.

[0004] However, the aforementioned existing technologies rely on fixed inventory thresholds for inventory warnings, making it impossible to dynamically adapt to changes in clinical dental needs and the consumption rate of materials themselves. This static warning system may trigger too late during peak demand or supply delays, potentially leading to shortages of critical materials; conversely, during periods of low demand, premature alerts can result in unnecessary inventory buildup, making misjudgments in inventory warnings highly likely. Summary of the Invention

[0005] This application provides an automated warehouse management method and system based on real-time monitoring and alarms. This method solves the problem that fixed thresholds in the prior art cannot cope with changes in the consumption rate and supply cycle fluctuations of dental materials, thereby improving the accuracy of dental medical material management.

[0006] Firstly, this application provides an automated silo management method based on real-time monitoring and alarms, applied to a silo management platform. The method includes: controlling an industrial robot to grab raw materials to be stored from a loading area based on grabbing instructions, and placing the raw materials in a detection area; after confirming that the raw materials are in the detection area, detecting the raw materials to obtain raw material attribute information, and determining the raw material type based on the raw material attribute information; determining a target silo among multiple silos based on the raw material attribute information and silo allocation rules, controlling the industrial robot to move the raw materials to be stored from the detection area into the target silo, and updating the inventory corresponding to the raw material type in real time based on the receiving action, and updating the storage status of the target silo; when the inventory of the target raw material type changes, obtaining data from all silos... The system retrieves the real-time inventory level corresponding to the target raw material type, and obtains historical demand data, procurement cycle data, and average daily consumption data for that target raw material type, which can be any raw material type in the warehouse. It calculates the demand standard deviation and procurement cycle standard deviation based on the historical demand data and procurement cycle data. It then calculates the dynamic safety stock threshold for the target raw material type based on the average daily consumption, demand standard deviation, and procurement cycle standard deviation. The system compares the real-time inventory level with the dynamic safety stock threshold. When the real-time inventory level is less than or equal to the dynamic safety stock threshold, it automatically generates a replenishment warning for the target raw material type and sends this warning to the terminal device of the responsible person via communication, enabling them to initiate the replenishment process.

[0007] By adopting the above technical solution, the industrial robot realizes automated warehousing by picking up raw materials to be stored from the loading area and placing them in the detection area. After confirming that the raw materials to be stored are located in the detection area, the raw materials to be stored are subjected to multi-dimensional detection to obtain raw material attribute information, and the raw material type is determined based on the raw material attribute information. Based on the raw material attribute information and the storage location allocation rules, the target storage location is determined from multiple storage locations in the silo, and the industrial robot is controlled to accurately move the raw materials to be stored from the detection area into the target storage location. At the same time as the warehousing action is completed, the inventory quantity corresponding to the raw material type and the storage status of the target storage location are updated in real time to ensure the timeliness and accuracy of inventory data. When the inventory quantity of the target raw material type changes, the real-time inventory quantity corresponding to the target raw material type is automatically obtained from all storage locations. Historical demand data, procurement cycle data and average daily consumption corresponding to the target raw material type are obtained. The historical demand data, procurement cycle data and average daily consumption are comprehensively calculated to obtain a dynamic safety stock threshold, which solves the problem that the fixed threshold in the existing technology cannot cope with the changes in the consumption rate of dental materials and the fluctuation of the supply cycle. The system compares real-time inventory levels with dynamic safety stock thresholds in real time. When the real-time inventory level is less than or equal to the dynamic safety stock threshold, it automatically generates a replenishment warning for the target raw material type and sends the warning to the terminal device of the responsible person in a timely manner via communication. This allows the responsible person to initiate the replenishment process in advance, effectively avoiding the risk of supply disruption of key materials due to late warnings during peak demand periods. At the same time, it also avoids inventory backlog caused by premature alarms during periods of low demand, significantly improving the accuracy of oral medical material warehouse management.

[0008] Optionally, the dynamic safety stock threshold for the target raw material type is calculated by analyzing the average daily consumption, demand standard deviation, and procurement cycle standard deviation. Specifically, this includes: calculating the average procurement cycle from the procurement cycle data; determining the service level coefficient based on the importance of the target raw material type; multiplying the average procurement cycle by the square of the demand standard deviation to obtain the demand volatility variance contribution; multiplying the square of the average daily consumption by the square of the procurement cycle standard deviation to obtain the procurement cycle volatility variance contribution; adding the demand volatility variance contribution and the procurement cycle volatility variance contribution to obtain the total volatility variance; and calculating the square root of the total volatility variance to obtain the comprehensive volatility standard deviation; multiplying the service level coefficient by the comprehensive volatility standard deviation to obtain the safety stock quantity; multiplying the average daily consumption by the average procurement cycle to obtain the average cycle consumption; and adding the average cycle consumption to the safety stock quantity to obtain the dynamic safety stock threshold.

[0009] By adopting the above technical solution, the average procurement cycle is calculated from the procurement cycle data, and the service level coefficient is determined based on the importance of the target raw material type. The contribution of demand fluctuation variance is obtained by multiplying the average procurement cycle by the square of the demand standard deviation, and the contribution of procurement cycle fluctuation variance is obtained by multiplying the square of the average daily consumption by the square of the procurement cycle standard deviation. The contribution of demand fluctuation variance and the contribution of procurement cycle fluctuation variance are added to obtain the total fluctuation variance, and the square root of the total fluctuation variance is calculated to obtain the comprehensive fluctuation standard deviation. The service level coefficient is multiplied by the comprehensive fluctuation standard deviation to obtain the safety stock calculation method. This ensures that high-importance materials can obtain a higher safety stock buffer, while low-importance auxiliary consumables maintain a lower inventory level, thereby optimizing inventory costs while ensuring the continuity of clinical treatment. The average daily consumption is multiplied by the average procurement cycle to obtain the average cycle consumption, and the average cycle consumption is added to the safety stock to obtain the dynamic safety stock threshold. This solves the problem in the existing technology where the fixed threshold causes supply disruptions due to late warnings during peak demand periods and stockpiles due to early warnings during periods of low demand. This effectively reduces the risk of shortages and inventory holding costs of oral medical materials.

[0010] Optionally, the service level coefficient is determined based on the importance of the target raw material type, specifically including: obtaining multiple importance assessment factors associated with the target raw material type; weighting the multiple importance assessment factors to obtain a comprehensive importance index, where the importance assessment factors include clinical criticality level, customization level, and supply scarcity level; comparing the comprehensive importance index with a first importance threshold and a second importance threshold, where the second importance threshold is greater than the first importance threshold; when the comprehensive importance index is less than the first importance threshold, classifying the target raw material type into a low importance interval, and calculating the target service level percentage based on a first preset linear function; when... When the overall importance index is greater than or equal to the first importance threshold and less than the second importance threshold, the target raw material type is classified into the medium importance interval, and the target service level percentage is calculated based on the second preset linear function, the slope of which is greater than the slope of the first preset linear function. When the overall importance index is greater than or equal to the second importance threshold, the target raw material type is classified into the high importance interval, and the target service level percentage is calculated based on the preset exponential function, wherein the target service level percentage increases at an accelerated rate as the overall importance index increases. The service level coefficient is obtained by calculating the target service level percentage based on the inverse cumulative distribution function.

[0011] By adopting the above technical solution, multiple importance assessment factors associated with the target raw material type, including clinical criticality level, customization level, and supply scarcity level, are obtained. A comprehensive importance index is calculated by weighting these multiple importance assessment factors. The comprehensive importance index is compared with the first importance threshold and the second importance threshold. Based on the comparison results, the corresponding calculation method is selected to determine the target service level percentage. The service level coefficient is calculated based on the inverse cumulative distribution function, and the service level target is transformed into the statistical parameters required for safety stock calculation. This allows materials of different importance to show significant differences in the dynamic safety stock threshold calculation, effectively balancing the shortage risk control of critical materials with the inventory cost optimization of ordinary materials.

[0012] Optionally, after updating the inventory quantity corresponding to the raw material type in real time based on the inbound action and updating the storage status of the target silo, the method further includes: after confirming that the raw material to be stored is located in the target silo, updating the real-time storage status associated with the target silo to the occupied status; obtaining the unique identifier of the target silo, associating the unique identifier with the raw material attribute information, and storing the mapping relationship in the silo status database; reading the real-time storage status of each silo in the silo, if the real-time storage status of the current silo is occupied, controlling the status indicator light corresponding to the current silo to display the first color; if the real-time storage status of the current silo is idle, controlling the status indicator light corresponding to the current silo to display the second color, the first color and the second color being different.

[0013] By adopting the above technical solution, once the raw materials to be stored are confirmed to be located in the target silo, the real-time storage status associated with the target silo is immediately updated to an occupied status, achieving instant synchronization of silo status and avoiding duplicate allocation errors caused by delayed silo status updates. The unique identifier of the target silo is obtained and mapped to the raw material attribute information, establishing a precise correspondence between silos and materials. Storing this mapping relationship in the silo status database ensures the timely preservation of silo-material association information, enabling rapid location of storage locations for materials with specific attributes. By reading the real-time storage status of each silo in the warehouse, and controlling the corresponding status indicator light to display different colors based on the current real-time storage status of the silo, visual management of the storage status is achieved, allowing warehouse managers to clearly understand the space utilization of the entire warehouse.

[0014] Optionally, the target storage location is determined from multiple storage locations in the silo based on raw material attribute information and storage location allocation rules. Specifically, this includes: obtaining the raw material type, usage frequency, and chemical compatibility list of the raw material to be stored; obtaining the real-time status and access path distance of all storage locations in the silo, where the real-time status includes occupied or idle status, and the access path distance is the path distance from the starting point of the industrial robot to each storage location; filtering all storage locations that are idle in real-time to obtain an initial set of available storage locations; obtaining the adjacent storage locations of the idle storage locations from the initial set of available storage locations, and if the raw material already stored in the adjacent storage location is marked as incompatible with the raw material to be stored in the chemical compatibility list, then the idle storage location is removed from the initial set of available storage locations to obtain a set of candidate storage locations, where the idle storage location is any one of the storage locations in the initial set of available storage locations; and determining the target storage location from the candidate storage location set based on usage frequency.

[0015] By adopting the above technical solution, the raw material type, usage frequency, and chemical compatibility list of the raw materials to be stored are obtained. The real-time status and access path distance of all silos in the silo are obtained. The silos in the real-time idle status are filtered to obtain an initial set of available silos, effectively excluding silos that are already occupied. The adjacent silos of each idle silo are obtained from the initial set of available silos, and it is checked whether the raw materials stored in the adjacent silos are marked as incompatible on the chemical compatibility list of the raw materials to be stored. If chemical incompatibility is detected, the idle silo is removed from the initial set of available silos to obtain a set of candidate silos. The target silo is determined from the set of candidate silos based on the usage frequency. This shortens the average moving distance of the industrial robot in daily storage and retrieval operations and improves the overall operating efficiency of the warehousing system.

[0016] Optionally, the target storage location is determined from the candidate storage location set based on usage frequency. Specifically, this includes: dividing all storage locations in the silo into high-frequency, medium-frequency, and low-frequency regions based on access path distance; when the usage frequency is greater than a preset first frequency, selecting storage locations located in the high-frequency region from the candidate storage location set to obtain a high-frequency candidate subset, and selecting the storage location with the smallest access path distance from the high-frequency candidate subset as the target storage location; if there are no candidate storage locations in the high-frequency region, selecting the storage location with the smallest access path distance from the candidate storage locations in the medium-frequency and low-frequency regions as the target storage location; when the usage frequency is less than or equal to the preset first frequency and greater than a preset second frequency, selecting the storage location located in the medium-frequency and low-frequency regions from the candidate storage location set... For positions in the mid-frequency region, a mid-frequency candidate subset is obtained, with a preset first frequency greater than a preset second frequency. Positions with the smallest access path distance are selected from the mid-frequency candidate subset as target positions. If no candidate positions exist in the mid-frequency region, positions with the smallest access path distance are selected from the candidate positions in the high-frequency and low-frequency regions as target positions. When the frequency used is less than or equal to the preset second frequency, positions located in the low-frequency region are selected from the candidate position set to obtain a low-frequency candidate subset. Positions with the largest access path distance are selected from the low-frequency candidate subset as target positions. If no candidate positions exist in the low-frequency region, positions with the largest access path distance are selected from the candidate positions in the mid-frequency and high-frequency regions as target positions.

[0017] By adopting the above technical solution, all storage locations in the silo are scientifically divided according to the access path distance to obtain high-frequency, medium-frequency, and low-frequency areas. Based on the usage frequency, suitable storage locations are selected from the high-frequency, medium-frequency, and low-frequency areas, so that materials with high usage frequency are assigned to storage locations with shorter access path distances, while materials with low usage frequency are assigned to storage locations with longer access path distances, thereby optimizing the overall storage efficiency of the warehousing system.

[0018] Optionally, the raw material to be stored is inspected to obtain its attribute information, specifically including: scanning the raw material to be stored using a laser scanner to obtain three-dimensional point cloud data; calculating the three-dimensional point cloud data to obtain the size information of the raw material to be stored; determining the thickness information based on the size information; acquiring side and top color images of the raw material to be stored; performing color space analysis on the top color image to obtain color information; segmenting and extracting the side color image to obtain the raw material color image; performing signal conversion on the raw material color image to obtain multiple projection signals; calculating the derivatives of the multiple projection signals to obtain multiple gradients; calculating the peak values ​​from the multiple gradients to obtain peak coordinates; and outputting the peak coordinates as layering information; measuring the light intensity value after light penetrates the raw material to be stored and obtaining the initial light intensity value before the raw material to be stored is placed; using the ratio of the light intensity value to the initial light intensity value as the transmittance; and outputting the size information, thickness information, color information, layering information, and transmittance as raw material attribute information.

[0019] By adopting the above technical solution, a laser scanner is used to scan the raw material to be stored to obtain three-dimensional point cloud data. The size information of the raw material to be stored is calculated from the three-dimensional point cloud data, and the thickness information is determined based on the size information. Side and top color images of the raw material to be stored are acquired. Color information is obtained by performing color space analysis on the top color image. The side color image is segmented and extracted to obtain the raw material color image. The raw material color image is then converted into multiple projection signals. Multiple gradients are obtained by calculating the derivatives of the multiple projection signals. Peak coordinates are obtained by calculating the peak values ​​of the multiple gradients and output as layering information. The light intensity value after light penetrates the raw material to be stored is measured, and the initial light intensity value before the raw material is placed is obtained. The ratio of the light intensity value to the initial light intensity value is used as the transmittance. The size information, thickness information, color information, layering information, and transmittance are output as raw material attribute information, providing data support for subsequent intelligent warehouse allocation and accurate inventory management.

[0020] The second aspect of this application provides an automated silo management system based on real-time monitoring and alarms. The system is a silo management platform, which includes a loading unit, a detection unit, a processing unit, and an alarm unit. The loading unit controls an industrial robot to pick up raw materials to be stored from the loading area based on grasping instructions and places them in the detection area. The detection unit, after confirming that the raw materials are in the detection area, detects them to obtain their attribute information and determines the material type based on this information. The processing unit, based on the raw material attribute information and silo allocation rules, determines the target silo from multiple silos in the silo, controls the industrial robot to move the raw materials from the detection area into the target silo, and updates the inventory corresponding to the material type in real time based on the receiving action, updating the storage status of the target silo. When the inventory level of a target raw material changes, the system retrieves the real-time inventory level of that target raw material type from all warehouses, along with historical demand data, procurement cycle data, and average daily consumption data corresponding to that target raw material type. The target raw material type can be any raw material type in any warehouse. The system calculates the demand standard deviation and procurement cycle standard deviation based on the historical demand data and procurement cycle data. It then calculates the dynamic safety stock threshold for the target raw material type based on the average daily consumption, demand standard deviation, and procurement cycle standard deviation. The system compares the real-time inventory level with the dynamic safety stock threshold. An alarm unit automatically generates a replenishment warning for the target raw material type when the real-time inventory level is less than or equal to the dynamic safety stock threshold. This warning is then sent to the terminal device of the responsible party via communication, enabling them to initiate the replenishment process.

[0021] In a third aspect, this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory, causing the electronic device to perform any of the methods described above in this application.

[0022] In a fourth aspect, this application provides a computer-readable storage medium storing instructions that, when executed, perform any of the methods described above in this application.

[0023] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Industrial robots automate the warehousing process by picking up raw materials to be stored from the loading area and placing them in the inspection area. After confirming that the raw materials are located in the inspection area, the robots perform multi-dimensional inspections to obtain raw material attribute information and determine the raw material type based on this information. Based on the raw material attribute information and the storage location allocation rules, the robot selects the target storage location from multiple storage locations in the warehouse and precisely moves the raw materials from the inspection area to the target storage location. Simultaneously, the robot updates the inventory quantity corresponding to the raw material type and the storage status of the target storage location in real time, ensuring the timeliness and accuracy of the inventory data. When the inventory quantity of the target raw material type changes, the robot automatically retrieves the real-time inventory quantity corresponding to that target raw material type from all warehouses, along with historical demand data, procurement cycle data, and average daily consumption data. A dynamic safety stock threshold is obtained by comprehensively calculating these data, solving the problem that fixed thresholds in existing technologies cannot cope with changes in the consumption rate and supply cycle fluctuations of dental materials. The system compares real-time inventory levels with dynamic safety stock thresholds in real time. When the real-time inventory level is less than or equal to the dynamic safety stock threshold, it automatically generates a replenishment warning for the target raw material type and sends the warning to the terminal device of the responsible person in a timely manner via communication. This allows the responsible person to initiate the replenishment process in advance, effectively avoiding the risk of supply disruption of key materials due to late warnings during peak demand periods. At the same time, it also avoids inventory backlog caused by premature alarms during periods of low demand, significantly improving the accuracy of oral medical material warehouse management. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating an automated silo management method based on real-time monitoring and alarm provided in an embodiment of this application; Figure 2 This is a schematic diagram of a scenario for an automated silo management method based on real-time monitoring and alarm provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0025] Explanation of reference numerals in the attached figures: 300, electronic device; 301, processor; 302, memory; 303, user interface; 304, network interface; 305, communication bus. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0027] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0028] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0029] Therefore, how to address the inability of fixed thresholds in existing technologies to cope with changes in the consumption rate and supply cycle fluctuations of dental materials is a pressing issue. This application provides an automated silo management method based on real-time monitoring and alarms, applied in a silo management platform. This silo management platform provides material management services for medical dental materials. Figure 1 This is a flowchart illustrating an automated silo management method based on real-time monitoring and alarm provided in an embodiment of this application. (Refer to...) Figure 1 The method includes the following steps S101-S109.

[0030] S101: Based on the grasping command, the industrial robot is controlled to grasp the raw materials to be stored from the feeding area and place the raw materials to be stored in the detection area.

[0031] In step S101 above, the system control unit sends a gripping command to the industrial robot. This command includes the spatial coordinates of the loading area and the position data of the material to be gripped. Upon receiving the gripping command, the industrial robot's robotic arm moves from its initial position to the loading area according to a preset trajectory, gripping the material using a vacuum suction cup or pneumatic gripper mounted on its end effector. Since medical dental materials such as zirconia ceramic blocks and resin blocks typically have regular geometric shapes and smooth surfaces, the vacuum suction cup provides stable suction force. For larger or irregularly shaped materials, pneumatic grippers are used for multi-point clamping to ensure that the material does not slip or shift position during gripping. Medical dental materials refer to various professional materials used for dental restoration, treatment, and cosmetic purposes, including zirconia ceramic blocks, glass-ceramic blocks, titanium alloy implants, silicone rubber impression materials, dentures, and resin-based composite materials.

[0032] After successfully grasping the raw material to be stored, the industrial robot smoothly moves the material from the loading area to the inspection area according to a pre-planned path. The inspection area is located in front of the hopper entrance and is equipped with a fixed inspection platform and lighting. The surface of the inspection platform uses a high-contrast background material to facilitate image acquisition by the vision sensor, and the lighting provides a stable and uniform light source environment to eliminate the influence of ambient light fluctuations on the inspection results. The industrial robot gently places the raw material to be stored in the center of the inspection platform and confirms that the placement action has been completed and the material is in a stable state through feedback from a force sensor. By setting up the inspection area independently rather than inspecting within each hopper, centralized configuration and unified management of the inspection equipment can be achieved, avoiding the increased cost and maintenance complexity caused by repeatedly deploying sensor equipment in multiple hoppers. Those skilled in the art will understand that the industrial robot is a specific example of an automated handling device, and other types of automated handling devices, such as gantry robots and AGVs, are also applicable.

[0033] S102: After confirming that the raw material to be stored is located in the detection area, the raw material to be stored is detected to obtain raw material attribute information, and the raw material type is determined based on the raw material attribute information.

[0034] In the above S102, the detection area is equipped with a position confirmation sensor. When the industrial robot places the raw material to be stored on the detection platform, the position confirmation sensor detects that the raw material to be stored has entered the effective range of the detection area through an infrared grating or a proximity switch, thus triggering the detection process.

[0035] The process of detecting raw materials to be stored to obtain their attribute information includes: scanning the raw materials to be stored using a laser scanner to obtain 3D point cloud data; calculating the size information of the raw materials to be stored based on the 3D point cloud data; determining the thickness information based on the size information; acquiring side and top color images of the raw materials to be stored; performing color space analysis on the top color image to obtain color information; segmenting and extracting the side color image to obtain the raw material color image; performing signal conversion on the raw material color image to obtain multiple projection signals; calculating the derivatives of the multiple projection signals to obtain multiple gradients; calculating the peak values ​​from the multiple gradients to obtain peak coordinates; and outputting the peak coordinates as layering information; measuring the light intensity value after light penetrates the raw materials to be stored and obtaining the initial light intensity value before the raw materials are placed; and using the ratio of the light intensity value to the initial light intensity value as the transmittance; and outputting the size information, thickness information, color information, layering information, and transmittance as raw material attribute information.

[0036] Specifically, the laser scanner's emitting unit begins operation, generating a red laser beam with a wavelength of 635 nanometers. This beam, after passing through an optical shaping system, forms a linear light band that scans the surface of the material to be stored at a frequency of 100 times per second. The laser scanner's rotary motor drives the reflector assembly to rotate along a preset scanning path, ensuring the linear laser beam covers the entire surface area of ​​the material. When the laser beam strikes the surface, diffuse reflection occurs at different locations. The laser receiver captures the reflected light signal using a CMOS image sensor and calculates the distance from each measurement point to the scanner's reference plane based on triangulation principles. For a standard-sized zirconia block, the laser scanner can acquire over 500,000 spatial coordinate points within three seconds. These discrete coordinate points constitute point cloud data describing the three-dimensional morphology of the material. The three-dimensional point cloud data is imported into a point cloud processing algorithm module. This module performs noise reduction on the original point cloud to eliminate random error points during the scanning process. A statistical filtering method is used to calculate the average distance from each point to its neighbors, identifying and removing discrete points exceeding the standard deviation range as noise. After noise reduction, the point cloud data is fitted into a continuous 3D surface model using a surface reconstruction algorithm. Key geometric features of the raw material to be stored are then identified on this 3D surface model. For rectangular zirconia blocks, a planar detection method is used to identify six rectangular faces and calculate the distances between opposite faces, obtaining three dimensional parameters: length, width, and height. For cylindrical resin blocks, the algorithm identifies the boundary points of the circular cross-section and fits the circumference, calculating the diameter of the circle as dimensional information. Simultaneously, the axial distance between the two end faces of the cylinder is measured as height information.

[0037] Based on the acquired dimensional information, the thickness information is further determined. For multi-layered zirconia blocks, the thickness information usually corresponds to the dimension in the height direction of the zirconia block, while for thin-film glass-ceramic materials, the thickness information corresponds to the measurement value in the minimum dimension direction of the glass-ceramic material. This dimensional measurement method based on three-dimensional point cloud data can achieve an accuracy of 0.01 millimeters, which can meet the precision processing requirements of dental restorative materials.

[0038] The dual-camera system above the inspection platform simultaneously acquires images. The top camera uses a 5-megapixel color CMOS sensor with a 16mm lens focal length. The camera is mounted perpendicular to the platform and 300mm from its surface to ensure a complete image of the top surface of the material to be stored. A ring of LEDs is evenly distributed around the lens, emitting white light with a color temperature of 5500 Kelvin to eliminate interference from shadows for color recognition. Once triggered, the top camera captures a color image of the top surface, with the material to be stored occupying approximately 70% of the central area. The side camera is mounted to the side of the inspection platform, with its lens optical axis at a 45-degree angle to the platform. This angle allows observation of the material's side profile while avoiding measurement errors caused by perspective distortion. The side camera also features an independent linear light source, positioned along the side of the material to highlight edge contours and internal layered structures.

[0039] The top-view color image is converted from the camera's original RGB color space to the LAB color space. The LAB color space decomposes color information into a luminance channel (L), a red-green chromatic channel (A), and a yellow-blue chromatic channel (B), a color representation that more closely resembles the color perception characteristics of the human eye. The color analysis algorithm automatically identifies the effective area of ​​the material to be stored in the top-view color image, eliminating interference from background and edge reflective areas. Within the effective area, it extracts the LAB values ​​of all pixels and calculates the average and standard deviation to obtain color parameters representing the overall color characteristics of the material. The measured LAB values ​​are matched against a database of standard color charts for dental materials. This database stores standard LAB values ​​for internationally recognized color systems such as the VITA color chart and the 3D-MASTER colorimetric system. By calculating the color difference between the tested material and each standard color number, a color match is considered to occur when the color difference is less than three units, and the corresponding color information, such as A2 or 2M2, is output.

[0040] After the lateral color image enters the image segmentation module, a deep learning-based semantic segmentation algorithm classifies different regions in the image. This algorithm, trained on thousands of dental material sample images, can accurately identify the material body, detection platform, background, and reflective areas. The segmentation algorithm assigns a category label to each pixel, extracts pixels belonging to the material body to form a mask image, and then performs a logical AND operation between the original lateral color image and the mask image to obtain a result containing only the material color image. Pixel values ​​in the background region of the material color image are set to zero, while the original color information of the material region is preserved. Signal conversion processing is then performed on the material color image, scanning column by column along the vertical direction of the image and converting the color pixels in each column to grayscale values. The grayscale conversion formula can use a weighted average method. For a material color image with 1000 pixels in the height direction, the material color image is converted into 1000 grayscale projection signals, each representing the average grayscale intensity of the material at that vertical position. These projection signals form a grayscale curve that varies along the height direction. At the color transition interface of multiple material layers, due to the color difference between the upper and lower layers, the grayscale curve will show significant jumps.

[0041] The gradient signal is obtained by numerically differentiating multiple projected signals. The central difference method is used, where the gradient at the current point equals the difference between the previous and subsequent projected signal values, divided by two. This method effectively suppresses the influence of noise on gradient calculation. The gradient signal reflects the rate of change of the projected signals. At material layer interfaces, the gradient signal will exhibit positive or negative peaks. Positive peaks correspond to the transition from dark to light layers, while negative peaks correspond to the transition from light to dark layers. The peak detection algorithm searches for local extrema in the gradient signal sequence, setting the peak detection threshold to three times the standard deviation of the gradient signal. Only extrema exceeding the peak detection threshold are identified as valid peaks. For example, for a three-layer gradient zirconia ceramic block, two distinct peak coordinates can usually be detected. The vertical positions of these two peak coordinates in the image correspond to the interfaces between the first and second layers, and between the second and third layers, respectively. The detected peak coordinates are converted from image pixel coordinates to actual physical coordinates. Combined with the size information obtained from the previous laser scanning, the layer thickness is calculated. The peak coordinate sequence and the corresponding layer thickness value are output as layer information. This layer information provides an accurate reference for subsequent processing path planning.

[0042] The transmittance measurement device includes an LED light source located at the bottom of the detection platform and a light intensity sensor located above the material to be stored. The LED light source emits green light with a wavelength of 550 nanometers, which is located in the center of the visible spectrum and can well reflect the overall transmittance performance of the material. Before the measurement begins, the LED light source is controlled to be lit and stably emitted. The light intensity sensor measures the light intensity value after the light passes directly through the air and the transparent area of ​​the detection platform. This measured value is recorded as the initial light intensity value. For the measurement device used in this embodiment, the initial light intensity value is approximately 8,000 lux. After the industrial robot places the material to be stored in the transmittance measurement area of ​​the detection platform, the light emitted by the LED light source needs to pass through the detection platform, the material to be stored, and the air layer in sequence before reaching the light intensity sensor. Scattering, absorption, and reflection within the material cause light intensity attenuation. The light intensity sensor collects the light intensity value after penetrating the material. For high-transparency glass-ceramic materials for anterior teeth, the measured light intensity value is approximately 3,200 lux, while for opaque high-strength zirconia materials for posterior teeth, the light intensity value may be as low as 800 lux. The transmittance parameter is obtained by calculating the ratio of the light intensity value to the initial light intensity value. The transmittance of the aforementioned glass ceramic is 0.4, or 40%, and the transmittance of high-strength zirconia is 0.1, or 10%. The size and thickness information obtained from laser scanning, the color information obtained from top surface image analysis, the layering information obtained from side image processing, and the transmittance obtained from optical measurements are summarized, and these attribute parameters are encapsulated into raw material attribute information for output.

[0043] Furthermore, after detecting the raw material attribute information of the material to be stored, the raw material attribute information is matched with the standard parameter templates stored in the raw material type database. When the similarity between the attribute information of the raw material to be stored and the feature vector of a certain type of material exceeds a preset threshold, the raw material type of the raw material to be stored is determined. In addition to using the aforementioned equipment to detect the raw material to be stored, other non-contact optical sensors can also be used to scan the raw material to be stored to obtain its three-dimensional dimensions and surface optical properties. The specific equipment selected for detecting the raw material to be stored can be set based on the actual situation, and no further restrictions are imposed here.

[0044] S103: Based on raw material attribute information and storage location allocation rules, determine the target storage location among multiple storage locations in the silo, control the industrial robot to move the raw material to be stored from the detection area into the target storage location, and update the inventory quantity corresponding to the raw material type in real time based on the warehousing action, and update the storage status of the target storage location.

[0045] In S103 above, after receiving the raw material attribute information of the raw material to be stored, the target storage location is determined from multiple storage locations in the silo based on the raw material attribute information and storage location allocation rules. Specifically, this includes: obtaining the raw material type, usage frequency, and chemical compatibility list of the raw material to be stored; obtaining the real-time status and access path distance of all storage locations in the silo, where the real-time status includes occupied or idle status, and the access path distance is the path distance from the starting point of the industrial robot to each storage location; filtering all storage locations with an idle status in real-time to obtain an initial set of available storage locations; obtaining the adjacent storage locations of the idle storage locations from the initial set of available storage locations, and if the raw material already stored in the adjacent storage location is marked as incompatible with the raw material to be stored in the chemical compatibility list, then the idle storage location is removed from the initial set of available storage locations to obtain a candidate storage location set, where the idle storage location is any one of the storage locations in the initial set of available storage locations; and determining the target storage location from the candidate storage location set based on the usage frequency.

[0046] Specifically, after obtaining the raw material attribute information, the raw material type is determined based on this information. For example, if the raw material attribute information shows dimensions of 14mm×18mm×20mm, color number A2, light transmittance of 15%, and a single-layer structure, the feature matching algorithm locates the raw material to be stored in the raw material type database as belonging to the standard zirconia block category. All outbound events for A2 color zirconia blocks within the past 90 days are extracted from the outbound history record table. Statistics show that this raw material type was used 126 times during this period, resulting in an average usage frequency of 9.8 times per week. The chemical compatibility list was retrieved from the Materials Safety Management Database, which stores an interaction matrix of 48 commonly used dental restorative materials in XML format. For zirconia ceramic blocks, the compatibility record shows that they are incompatible with PMMA resin. This is because PMMA resin slowly releases residual monomer molecules at room temperature. These organic compounds with a molecular weight of about 100 Daltons can diffuse in the air. When the concentration reaches 5 ppm, they will form an organic contamination layer on the zirconia surface, resulting in a 40% to 60% reduction in the bonding strength between the adhesive and the ceramic block surface during subsequent bonding. Zirconia ceramic blocks are marked as compatible with lithium disilicate glass ceramics, and marked as neutral with cobalt-chromium alloys, indicating that the two can be stored adjacently without producing a synergistic effect.

[0047] A monitoring network consisting of 48 pressure sensors and infrared beam sensors collects real-time data on the bin status. The bins are arranged in a 6x8 matrix layout, with each bin measuring 300mm × 300mm × 400mm. The bottom pressure sensors use strain gauge weighing units with an accuracy of 10g, continuously monitoring the bin load with a sampling period of 200ms. When the total weight of material in a bin exceeds 50g, the real-time status field of that bin is updated to "occupied," and the occupancy timestamp, material type ID, and current storage quantity are recorded in the database. When the weight drops below 50g, the status switches to "idle." A scan of the real-time status data table at the current query time revealed that 16 of the 48 bins were idle. Spatial distance calculations are then performed based on the kinematic model of the industrial robot. The starting point of the industrial robot's operation is defined as the center point of the detection platform, and the coordinates of the center point are obtained. Then, the entrance center coordinates of each bin in the bin are sequentially obtained. The 3D Euclidean distance formula is used to calculate the path distance between each bin and the industrial robot, and a bin path distance index table is established for subsequent queries.

[0048] The idle positions are selected and grouped into an initial set of available positions. Then, the adjacent positions of each position in the initial set of available positions are retrieved. Each position has a maximum of four adjacent positions, located in the top, bottom, left, and right directions. Positions located at the edges will have fewer adjacent positions. Figure 2As shown. After obtaining adjacent storage locations, if no raw materials are stored in an adjacent storage location, it is removed from the adjacent storage location to ensure that all obtained adjacent storage locations are already storing raw materials. The target storage location is obtained from the initial set of available storage locations. The chemical compatibility list of the raw materials to be stored is matched with the raw materials already stored in the target adjacent storage location. The target adjacent storage location and the target storage location are adjacent to each other. If they are marked as incompatible, the target storage location is immediately removed from the initial set of available storage locations. For example, storage location 3 is taken from the initial set of available storage locations for compatibility checking. Storage location 3 is located in the first row and third column, and its adjacent storage locations include storage locations 2, 4, and 13. Querying the stored material types of these three adjacent storage locations reveals that storage location 2 stores B1 color zirconia blocks, storage location 4 is empty, and storage location 13 stores light-cured resin blocks. The chemical compatibility list of the A2 color zirconia ceramic blocks to be stored was matched with the UV-cured resin blocks in compartment 13. The query results showed that the two were marked as incompatible, so a removal operation was immediately performed, and compartment 3 was removed from the initial available compartment set. The same compatibility check was then performed on the other compartments in the initial available compartment set. For compartment 15, the adjacent compartments 14, 16, 5, and 25 respectively stored A3 color zirconia ceramic blocks, glass-ceramic veneers, an empty compartment, and A1 color zirconia ceramic blocks. These materials were all marked as compatible or neutral with the zirconia ceramic blocks to be stored, so compartment 15 was retained in the initial available compartment set. After all compatibility checks, the remaining compartments in the initial available compartment set were used to form a candidate compartment set. Chemical compatibility screening can effectively avoid cross-contamination between different materials and ensure that the raw materials to be stored maintain their original performance without being affected during storage.

[0049] Furthermore, the target storage location is determined from the candidate storage location set based on usage frequency. Specifically, this includes: dividing all storage locations in the silo into high-frequency, medium-frequency, and low-frequency regions according to the access path distance; when the usage frequency is greater than a preset first frequency, storage locations located in the high-frequency region are selected from the candidate storage location set to obtain a high-frequency candidate subset, and the storage location with the smallest access path distance is selected from the high-frequency candidate subset as the target storage location; if there are no candidate storage locations in the high-frequency region, the storage location with the smallest access path distance is selected from the candidate storage locations in the medium-frequency and low-frequency regions as the target storage location; when the usage frequency is less than or equal to the preset first frequency and greater than a preset second frequency, the storage location located in the medium-frequency and low-frequency regions is selected from the candidate storage location set to obtain a high-frequency candidate subset, and the storage location with the smallest access path distance is selected from the medium-frequency and low-frequency regions as the target storage location; when the usage frequency is less than or equal to the preset first frequency and greater than the preset second frequency, the storage location located in the medium-frequency and low-frequency regions is selected from the candidate storage location set to obtain a high-frequency candidate subset, and the storage location with the smallest access path distance is selected from the medium-frequency and low-frequency regions as the target storage location. For positions in the high-frequency region, a mid-frequency candidate subset is obtained, with a first preset frequency greater than a second preset frequency. Positions with the shortest access path distance are selected from the mid-frequency candidate subset as target positions. If no candidate positions exist in the mid-frequency region, positions with the shortest access path distance are selected from the candidate positions in the high-frequency and low-frequency regions as target positions. When the frequency used is less than or equal to the second preset frequency, positions located in the low-frequency region are selected from the candidate position set to obtain a low-frequency candidate subset. Positions with the longest access path distance are selected from the low-frequency candidate subset as target positions. If no candidate positions exist in the low-frequency region, positions with the longest access path distance are selected from the candidate positions in the mid-frequency and high-frequency regions as target positions.

[0050] Specifically, all storage locations in the silo are divided into regions based on the access path distance. The path distance index table is used to obtain the path distance data from each storage location to the starting point of the industrial robot. All distance values ​​are arranged in ascending order to form an ordered distance sequence. The storage locations corresponding to the first 20% of the distance values ​​in the ordered distance sequence are identified as high-frequency regions, the storage locations corresponding to the middle 30% of the distance values ​​are identified as medium-frequency regions, and the storage locations corresponding to the last 50% of the distance values ​​are identified as low-frequency regions.

[0051] For example, a silo has 48 storage locations. These locations are sorted from smallest to largest based on path distance. The number of locations in the high-frequency zone is 48 × 20% = 9.6. Rounding up, we get 10 locations. The 10 locations with the shortest distances are designated as the high-frequency zone, numbered 07, 11, 05, 03, 14, 19, 08, 16, 22, and 06. The number of locations in the medium-frequency zone is 48 × 30% = 14.4. Rounding down, we get 14 locations. The locations from the 11th to the 24th position after sorting are designated as the medium-frequency zone, including locations numbered 23, 12, 20, 15, 28, 26, 31, 18, 25, 35, 29, 32, 37, and 21. The low-frequency zone contains the remaining 24 storage locations. This distance-based three-level zone division reflects a hierarchical storage strategy in warehouse management, which can store materials in the most suitable locations according to their usage characteristics, thereby optimizing overall storage and retrieval efficiency.

[0052] Next, the usage frequency of the raw material to be stored is obtained and compared with a preset first frequency and a preset second frequency, where the preset first frequency is greater than the preset second frequency. When the usage frequency is greater than the preset first frequency, the raw material to be stored is confirmed to be a high-frequency usage category. Warehouses located in the high-frequency region are selected from the candidate warehouse set. By traversing each warehouse number in the candidate warehouse set and calculating the intersection of each warehouse number with the high-frequency region, a high-frequency candidate subset is obtained. Then, the warehouse with the smallest access path distance is selected from the high-frequency candidate subset and determined as the target warehouse.

[0053] In the example above, the raw material to be stored is used 9.8 times / week, and the preset first frequency is set to 8 times / week. Since 9.8 is greater than 8, the raw material to be stored belongs to the high-frequency usage category. The candidate warehouse set is {03, 05, 07, 12, 15, 18, 21, 27, 33, 39, 45}. Warehouses in the high-frequency area include 07, 11, 05, 03, 14, 19, 08, 16, 22, and 06. The intersection of these two sets yields the high-frequency candidate subset {07, 05, 03}. From these three warehouses, the warehouse with the shortest access path distance is selected. If the distance for warehouse 07 is 1856mm, the distance for warehouse 05 is 2134mm, and the distance for warehouse 03 is 2298mm, then warehouse 07 is determined as the target warehouse using a minimum value search algorithm. The strategy of prioritizing finding the shortest storage location in high-frequency areas ensures that frequently used materials are stored in the most convenient and quickest access locations, reducing the repetitive movement distance of the robotic arm. Based on 1.4 retrievals per day, compared to storing the material in storage location 45, which is farther away among the candidate storage locations, the daily movement distance can be reduced by 7467mm, and the cumulative monthly movement distance can be reduced by about 224 meters, significantly reducing equipment wear and energy consumption.

[0054] Furthermore, when the usage frequency is less than or equal to a preset first frequency, the usage frequency is compared with a preset second frequency. When the usage frequency is greater than the preset second frequency, it is determined that the raw material to be stored belongs to the medium-frequency usage category. Storage locations within the medium-frequency region are selected from the candidate storage location set, and a set intersection operation is performed to obtain a medium-frequency candidate subset. Then, the storage location with the smallest access path distance is selected from the medium-frequency candidate subset and designated as the target storage location. For example, if the raw material to be stored is B3 color glass-ceramic veneer, with a usage frequency of 5 times / week, and the preset second frequency is set to 3 times / week, a comparison is performed to determine that 5≤8 and 5>3 are true, thus determining that the raw material belongs to the medium-frequency usage category. Storage locations within the medium-frequency region are selected from the candidate storage location set to obtain a medium-frequency candidate subset {12, 15, 18, 21}, a total of 4 storage locations. The storage location with the shortest access path distance was selected from the mid-frequency candidate subset. Distance data showed that storage location 12 was 2561mm, storage location 15 was 2734mm, storage location 18 was 2912mm, and storage location 21 was 3156mm. Storage location 12 was selected as the target storage location based on the minimum distance. This strategy of allocating mid-frequency materials to the mid-frequency region ensures a certain level of access convenience while reserving optimal locations for high-frequency materials. It achieves rational tiered utilization of storage space and avoids the problem of low-frequency materials occupying nearby storage locations, which leads to a decrease in access efficiency for high-frequency materials.

[0055] When a special situation arises where no available candidate storage locations exist in a high-frequency region, a degraded allocation mechanism is activated. For example, suppose the raw material to be stored is A1 color zirconia ceramic blocks used 10 times per week. However, after chemical compatibility screening, all 10 storage locations in the high-frequency region are eliminated because adjacent storage locations store incompatible materials, resulting in an empty high-frequency candidate subset. An alternative plan is then implemented, selecting the storage location with the minimum access path distance from the candidate storage locations in the mid-frequency and low-frequency regions. Assuming the remaining candidate storage location set is {23, 26, 31, 40, 43, 46}, the distance values ​​for these storage locations are 2512mm, 2845mm, 2987mm, 3678mm, 3912mm, and 4234mm, respectively. Using a global minimum value search, storage location 23 with a distance of 2512mm is determined as the target storage location. Although slot 23 is located in the mid-frequency region, it still provides a relatively convenient access path for high-frequency materials because it is the shortest among the remaining candidate slots. Compared to random allocation to remote slots, this degradation strategy achieves suboptimal allocation under compatibility constraints.

[0056] Furthermore, when the usage frequency of the raw material to be stored is less than or equal to a preset second frequency, it is confirmed that the raw material belongs to the low-frequency usage type. Storage locations located in the low-frequency region are selected from the candidate storage location set to obtain a low-frequency candidate subset. Then, the storage location with the longest access path distance, rather than the shortest distance, is selected from the low-frequency candidate subset and used as the target storage location. For example, if the raw material to be stored is C4 color special dyed ceramic with a usage frequency of only 0.8 times / week, and the condition 0.8≤3 is met, it is determined that the raw material belongs to the low-frequency usage category. Storage locations located in the low-frequency region are selected from the candidate storage location set, resulting in a low-frequency candidate subset of 4 storage locations: {27, 33, 39, 45}. Instead of the minimum distance, the storage location with the longest access path was selected from the low-frequency candidate subset. The results showed that storage location 27 had a distance of 3345mm, 33 had a distance of 3789mm, 39 had a distance of 4012mm, and 45 had a distance of 4523mm. Storage location 45 was selected as the target location based on the maximum value. This reverse strategy of storing low-frequency materials in the furthest location saves valuable close-range storage space for high-frequency and mid-frequency materials. Since low-frequency materials are only accessed about 3 times a month, even if each access involves an additional 2667mm of movement (the difference from the nearest storage location), the total annual increase in movement is only 96 meters. This extra movement has a negligible impact on low-frequency materials, but it allows for the use of closer storage locations for high-frequency materials, optimizing the overall storage efficiency of the entire silo from a global perspective.

[0057] Furthermore, when no available candidate storage space is available in the mid-frequency region, the storage space with the shortest access path distance is selected from the candidate storage spaces in the high-frequency and low-frequency regions as the target storage space. When no available candidate storage space is available in the low-frequency region as well, the storage space with the longest access path distance is selected from the candidate storage spaces in the mid-frequency and high-frequency regions. For example, if the low-frequency candidate subset is empty and the remaining candidate storage spaces {05, 12, 18, 23} have query distances of 2134mm, 2561mm, 2912mm, and 2512mm respectively, storage space number 18, corresponding to the maximum value of 2912mm, is selected as the target storage space. This ensures that low-frequency materials preferentially occupy relatively far-away locations even under constraints, maintaining the basic principle of hierarchical storage.

[0058] After identifying the target storage location, a movement command is sent to the industrial robot. This command includes the target storage location's three-dimensional spatial coordinates and placement posture parameters. The robot's robotic arm moves from the detection area to the entrance of the target storage location along a planned collision-free path. The end effector, maintaining a stable gripping state, enters the storage location and places the material inside. Only one material can be placed in a single storage location. After the robot completes the placement, the end effector releases its grip. Once the force sensor detects that the gripping force has returned to zero, the robotic arm exits the storage location, and the robot returns with a storage completion signal. The system then queries the inventory data table for the record corresponding to the current material type and increments the inventory value for that material type. For example, if the initial inventory of A2 color zirconia blocks was 42 pieces, the updated inventory will be 43 pieces. The storage status information of the target storage location is simultaneously updated from idle to occupied. This real-time inventory and storage status update mechanism ensures the accuracy and timeliness of the warehouse management system data.

[0059] In one possible implementation, after updating the inventory corresponding to the raw material type in real time based on the warehousing action and updating the storage status of the target silo, the method further includes: after confirming that the raw material to be stored is located in the target silo, updating the real-time storage status associated with the target silo to an occupied status; obtaining the unique identifier of the target silo, associating the unique identifier with the raw material attribute information, and storing the mapping relationship in the silo status database; reading the real-time storage status of each silo in the silo, and if the real-time storage status of the current silo is occupied, controlling the status indicator light corresponding to the current silo to display a first color; if the real-time storage status of the current silo is idle, controlling the status indicator light corresponding to the current silo to display a second color, wherein the first color and the second color are different.

[0060] Specifically, after the industrial robot completes the action of storing raw materials in the target hopper, a pressure sensor at the bottom of the hopper continuously monitors changes in the hopper's load. When the pressure sensor detects a change in the load of the target hopper and it remains stable for a certain period of time, it determines that the raw materials have been stably placed. An industrial camera installed above the hopper acquires images of the target hopper, performs edge detection and contour extraction on the acquired images, identifies the corresponding size and color, and then compares the identified size and color with the raw material attribute information to be stored. If the difference between the two is within a preset threshold, it is confirmed that the raw materials to be stored have been successfully placed in the target hopper. This confirmation method based on multi-sensor fusion can effectively avoid misjudgment by a single sensor and ensure the accuracy of inventory data. Immediately after confirming that the raw materials are in place, a status update process is initiated. By accessing the hopper status table in the hopper status database, the record row of the target hopper number is located. This record row contains fields such as hopper identifier, real-time storage status, update timestamp, load weight, and sensor flag. The real-time storage status of the target hopper is changed from the original idle state to the occupied state, the timestamp is updated synchronously, and the load weight is recorded.

[0061] The unique identifier of the target silo is read from the silo hardware configuration table. An association mapping record is created, binding the unique identifier to the raw material attribute information, constructing a composite data object containing the raw material type, size, usage frequency, storage timestamp, and chemical compatibility list. This mapping relationship is stored in the material mapping table of the silo status database, establishing a traceable association between the unique identifier and the raw material attribute information. This mapping mechanism allows for quick querying of detailed material information by silo number, or conversely, finding the silo containing a material type. For example, when an operator needs to find A2 color zirconia ceramic blocks, the location information and inventory quantity of silo 07 can be quickly returned, significantly improving warehouse management efficiency.

[0062] By reading all records in the silo status database, the real-time storage status field value of each silo is obtained. Traversing the record set, it is found that 32 silos, including 07, 01, 04, and 13, have a real-time storage field value indicating they are occupied, while 16 silos, including 05, 06, and 09, have an idle status. Control commands are sent to the programmable logic controller (PLC) in the silo control cabinet via the Modbus RTU protocol. The PLC sends PWM pulse width modulation signals to the RGB three-color LED status indicator lights of each silo through the digital output module (DO). For silo 07, since its real-time storage status is occupied, the status indicator light for this silo displays the first color, red. For silo 05, since its real-time storage status is idle, the status indicator light for this silo displays the second color, green. This color coding scheme is based on the high sensitivity and easy differentiation of red and green to the human eye. Operators can quickly identify the occupancy status of silos from 3 meters away without having to query the system database one by one. When operators need to manually store or retrieve materials, they can locate the available silos within 1 second by observing the indicator light color, which improves efficiency compared to traditional solutions.

[0063] S104: When the inventory of the target raw material type changes, obtain the real-time inventory of the target raw material type from all warehouses, and obtain the historical demand data, procurement cycle data and average daily consumption corresponding to the target raw material type. The target raw material type is any raw material type in the warehouse.

[0064] In step S104 above, after the industrial robot places the raw material to be stored into the target hopper, the inventory of the target raw material type changes through the hopper status database. The number of hoppers currently storing the target raw material type is counted by traversing the material mapping tables of all hoppers. Since each hopper can only store one raw material, the real-time inventory corresponding to the target raw material type is directly obtained. This inventory statistics method based on hopper occupancy status ensures data accuracy, as the occupancy status of each hopper corresponds one-to-one with the material, avoiding duplicate counting or omissions that may occur with traditional counting methods. For example, if the target raw material type is zirconia A2 color ceramic blocks, the number of hoppers storing zirconia A2 color ceramic blocks in all hoppers is counted. The query results show that 6 hoppers (07, 14, 22, 31, 38, and 42) store zirconia A2 color ceramic blocks, therefore the real-time inventory corresponding to the target raw material type is 6 blocks.

[0065] Historical demand data corresponding to the target raw material type was extracted from the historical data warehouse. The query time window was set to the most recent 90 days, yielding 82 outbound records for the target raw material type between October 23, 202x and January 22, 202x. Each record includes the outbound date, warehouse location number, user department, and usage description. Analysis of this historical demand data revealed that the average outbound frequency for this raw material on weekdays was 1.3 times / day, decreasing to 0.1 times / day on weekends and holidays. Simultaneously, the procurement management module was accessed to read the procurement cycle data corresponding to the target raw material type. The records showed that the complete cycle from initiating a purchase request to raw material warehousing was 12 working days, including 2 days for supplier confirmation, 6 days for production, 3 days for logistics and transportation, and 1 day for quality inspection and warehousing.

[0066] The average daily consumption is calculated based on historical demand data. An exponentially weighted moving average algorithm is used to process the data from the most recent 30 days. Outbound records from the most recent 7 days are assigned a weighting coefficient of 0.5, records from days 8 to 14 have a weighting coefficient of 0.3, and records from days 15 to 30 have a weighting coefficient of 0.2. The weighted calculation formula yields an average daily consumption of 0.91 units / day. The three key data points—real-time inventory of 6 units from all warehouses, a procurement cycle of 12 working days, and an average daily consumption of 0.91 units / day—are simultaneously pushed to the alarm unit so that the alarm unit can determine the inventory level of the target raw material type.

[0067] S105: Calculate the standard deviation of demand and the standard deviation of procurement cycle by calculating the historical demand data and procurement cycle data respectively.

[0068] In step S105 above, after obtaining historical demand data, to quantify the uncertainty of demand fluctuations, statistical analysis is performed on this historical demand data to calculate the demand standard deviation. For example, after obtaining 82 historical demand records for zirconia A2 color blocks within the last 90 days, the 82 outbound records are categorized by date. Statistics show that outbound activity occurred on 58 working days within those 90 days: 1 block on 26 days, 2 blocks on 19 days, 3 blocks on 8 days, 1 block on 3 days, and no outbound activity on 2 days. The arithmetic mean of these 58 valid demand samples is calculated to be 1.41 blocks / day. The standard deviation is then calculated using the formula. By calculating and summing the squared deviations daily, the variance is obtained as 0.68. Taking the square root yields a demand standard deviation of 0.82 blocks / day. This demand standard deviation reflects the fluctuation range of daily consumption for this type of raw material; a larger value indicates more unstable demand, requiring a higher safety stock buffer.

[0069] The standard deviation of the procurement cycle is calculated by analyzing the procurement cycle data. For example, 18 procurement records for this target raw material type within the most recent 12 months are extracted from the historical procurement cycle table. Each record includes the procurement request date, actual delivery date, and the total number of days in the procurement cycle. Statistical analysis reveals the following cycle distribution for these 18 procurements: 6 with a cycle of 12 days, 5 with 11 days, 4 with 13 days, 2 with 14 days, and 1 with 10 days. The average procurement cycle is calculated to be 12.17 days. The standard deviation formula is used to process the 18 sample data, calculating the squared deviation of each procurement cycle from the average. The sum of these deviations, divided by the sample size of 18, yields a variance of 1.24. Taking the square root, the standard deviation of the procurement cycle is 1.11 days. This standard deviation quantifies the uncertainty of supply chain delivery time. A larger standard deviation indicates poor supplier delivery stability, potentially affected by factors such as logistical delays, changes in production schedules, or raw material shortages. Therefore, a higher time buffer factor needs to be introduced into the safety stock calculation.

[0070] S106: Calculate the average daily consumption, demand standard deviation, and procurement cycle standard deviation to obtain the dynamic safety stock threshold for the target raw material type.

[0071] In S106 above, the average daily consumption, demand standard deviation, and procurement cycle standard deviation are calculated to obtain the dynamic safety stock threshold for the target raw material type. Specifically, this includes: calculating the average procurement cycle from the procurement cycle data; determining the service level coefficient based on the importance of the target raw material type; multiplying the average procurement cycle by the square of the demand standard deviation to obtain the demand volatility variance contribution; multiplying the square of the average daily consumption by the square of the procurement cycle standard deviation to obtain the procurement cycle volatility variance contribution; adding the demand volatility variance contribution and the procurement cycle volatility variance contribution to obtain the total volatility variance; and calculating the square root of the total volatility variance to obtain the comprehensive volatility standard deviation; multiplying the service level coefficient by the comprehensive volatility standard deviation to obtain the safety stock quantity; multiplying the average daily consumption by the average procurement cycle to obtain the average cycle consumption; and adding the average cycle consumption to the safety stock quantity to obtain the dynamic safety stock threshold.

[0072] Specifically, after obtaining the average daily consumption, demand standard deviation, and procurement cycle standard deviation, the procurement records for the target raw material type are first read from the procurement cycle table. The average number of procurement records is then calculated to obtain the average procurement cycle. For example, after obtaining the average daily consumption of zirconia A2 color ceramic blocks (0.91 blocks / day), demand standard deviation (0.82 blocks / day), and procurement cycle standard deviation (1.11 days), the most recent 18 procurement records for this target raw material type are read from the procurement cycle table. The cycle distribution of these 18 procurements is as follows: 6 times in 12 days, 5 times in 11 days, 4 times in 13 days, 2 times in 14 days, and 1 time in 10 days. The average procurement cycle is calculated using the arithmetic mean formula to be 12.17 days. This value represents the normal time required from initiating a procurement request to the raw material being stored in the warehouse, and it serves as the time base parameter for subsequently calculating the average cycle consumption.

[0073] The service level coefficient is determined based on the importance of the target raw material type in the production process, and is obtained by accessing the material importance table. The contribution of demand fluctuation variance is obtained by multiplying the average procurement cycle by the square of the demand standard deviation. For example, with an average procurement cycle of 12.17 days and a demand standard deviation of 0.82 units / day, the calculation formula is 12.17*(0.82). 2 =8.18, resulting in a demand fluctuation variance contribution of 8.18 blocks. This calculation quantifies the inventory uncertainty caused by daily demand fluctuations within a fixed procurement cycle. When the demand standard deviation is large, even with a stable procurement cycle, more inventory needs to be maintained to cope with sudden high demand situations. Simultaneously, the square of the average daily consumption is multiplied by the square of the procurement cycle standard deviation to obtain the procurement cycle fluctuation variance contribution. For example, with an average daily consumption of 0.91 blocks / day and a procurement cycle standard deviation of 1.11 days, the calculation formula (0.91) is used. 2 *(1.11) 2 =1.02, resulting in a procurement cycle fluctuation variance contribution of 1.02 blocks. This calculation process quantifies the inventory risk caused by unstable supplier delivery times. Even if daily demand is relatively stable, frequent delays in the procurement cycle can increase the risk of inventory depletion. For example, if a procurement is delayed to 14 days instead of the average 12.17 days, an additional 1.67 blocks of raw materials will be consumed during these extra 1.83 days, based on the average daily consumption.

[0074] The total variance is calculated by adding the contribution of demand volatility variance to the contribution of procurement cycle volatility variance. This total variance comprehensively reflects the combined impact of uncertainties on inventory management from both the demand and supply sides. For example, if the demand volatility variance contribution is 8.18 blocks and the procurement cycle volatility variance contribution is 1.02 blocks, the total variance is 8.18 + 1.02 = 9.20 blocks. Calculating the square root of the total variance yields a composite standard deviation of 3.03 blocks. This composite standard deviation represents the typical deviation between actual and expected consumption over the entire procurement cycle. Compared to considering demand or supply volatility separately, this combined calculation method more accurately assesses inventory risk because, in actual operations, demand volatility and procurement cycle volatility often coexist and overlap.

[0075] Multiplying the service level coefficient by the overall standard deviation of volatility yields the safety stock. For example, with a service level coefficient of 2.33 and an overall standard deviation of volatility of 3.03 blocks, the calculation is 2.33 * 3.03 ≈ 7.06, resulting in a safety stock of 7.06 blocks, rounded up to 8 blocks. This safety stock represents the amount of buffer inventory needed to achieve a 99% service level, in addition to the normal cycle consumption. When actual demand exceeds expectations or the procurement cycle is extended, this 8 blocks of safety stock can fill the gap and prevent production interruptions. Multiplying the average daily consumption by the average procurement cycle yields the average cycle consumption. For example, with an average daily consumption of 0.91 blocks / day and an average procurement cycle of 12.17 days, the calculation is 0.91 * 12.17 ≈ 11.07, resulting in an average cycle consumption of 11.07 blocks, rounded up to 12 blocks. This average cycle consumption represents the amount of raw materials consumed at the historical average rate within a complete procurement cycle and serves as the baseline demand for inventory management. Adding the average cycle consumption of 12 blocks to the safety stock of 8 blocks, we obtain a dynamic safety stock threshold of 20 blocks for this target raw material type. This threshold is stored in the inventory threshold table for subsequent comparison with the real-time inventory of the target raw material type.

[0076] Furthermore, in addition to determining the service level coefficient by querying the material importance table, the service level coefficient is determined based on the importance of the target raw material type. Specifically, this includes: obtaining multiple importance assessment factors associated with the target raw material type; weighting these factors to obtain a comprehensive importance index; the importance assessment factors include clinical criticality level, customization level, and supply scarcity level; comparing the comprehensive importance index with a first importance threshold and a second importance threshold, where the second importance threshold is greater than the first importance threshold; and when the comprehensive importance index is less than the first importance threshold, classifying the target raw material type into a low importance interval and calculating the service level coefficient based on a first preset linear function. The target service level percentage is calculated as follows: When the overall importance index is greater than or equal to the first importance threshold and less than the second importance threshold, the target raw material type is classified into the medium importance interval, and the target service level percentage is calculated based on the second preset linear function, the slope of which is greater than the slope of the first preset linear function; when the overall importance index is greater than or equal to the second importance threshold, the target raw material type is classified into the high importance interval, and the target service level percentage is calculated based on the preset exponential function, wherein the target service level percentage increases at an accelerated rate with the increase of the overall importance index; the service level coefficient is obtained by calculating the target service level percentage based on the inverse cumulative distribution function.

[0077] Specifically, the material importance factor table is first used to query multiple importance assessment factors associated with the target material type. This involves reading the clinical criticality level, customization level, and supply scarcity level of the target material type. The clinical criticality level indicates the visible location of the material used in anterior aesthetic restorations, directly impacting patient treatment outcomes and satisfaction. Material shortages will lead to appointment delays. The customization level indicates that the material is customized to match the specific patient's tooth color requirements and cannot be replaced by other shades; each order has unique and irreplaceable shade requirements. The supply scarcity level indicates that there are few suppliers of this material, and its production cycle is long, making it difficult to find emergency substitutes on the market, resulting in a high vulnerability of the supply chain. For example, if the target material type is zirconia A2 shade porcelain blocks, querying the material importance factor table for multiple importance assessment factors associated with zirconia A2 shade porcelain blocks reveals a clinical criticality level of 9, a customization level of 8, and a supply scarcity level of 7.

[0078] A weighted calculation was performed on these multiple importance assessment factors to obtain a comprehensive importance index. The weight coefficients for each factor were retrieved from the weight allocation table: the weight coefficient for clinical criticality level was 0.5, the weight coefficient for customization level was 0.3, and the weight coefficient for supply scarcity level was 0.2. These weight coefficients were set based on the application scenario of this warehousing management system in the dental prosthesis processing industry. Clinical criticality has the most direct impact on patient treatment outcomes and therefore has the highest weight. A weighted summation calculation yielded 8.3, meaning the comprehensive importance index for the target raw material type is 8.3 points. This comprehensive importance index integrates the assessment results of the three dimensions—clinical impact, customization characteristics, and supply chain risk—into a single quantitative indicator, facilitating unified threshold comparisons and service level decisions.

[0079] The first and second importance thresholds are read from the threshold configuration table. The first importance threshold is set to 5.0 points, and the second importance threshold is set to 8.0 points. These two thresholds divide the comprehensive importance index range of 0 to 10 points into three importance intervals: low, medium, and high. The calculated comprehensive importance index of 8.3 points is compared with the first importance threshold of 5.0 points. If the comprehensive importance index is greater than the first importance threshold, it is further compared with the second importance threshold. If the comprehensive importance index is also greater than the second importance threshold, the target raw material type is classified into the high importance interval. If the comprehensive importance index is less than the first importance threshold, the target raw material type is classified into the low importance interval. If the comprehensive importance index is greater than or equal to the first importance threshold and less than the second importance threshold, the target raw material type is classified into the medium importance interval.

[0080] Once the target raw material type is determined to belong to the high importance range, the parameters of the preset exponential function are read from the function configuration table. The expression of the preset exponential function is S(x)=85+15*(1-e (-0.5*(x-8)) S(x) represents the overall importance index, and S(x) represents the target service level percentage. This function is designed to ensure that when the overall importance index reaches 8.0, the service level percentage starts at 85%, and the service level increases at an accelerated rate as the importance index increases. Substituting the overall importance index of 8.3 into the preset index function, the calculation is S(8.3) = 85 + 15 * (1 - e^(-x / x))^(-x / x). (-0.5*(8.3-8)) The overall importance index of the raw material is 87.09, resulting in a target service level percentage of approximately 87.09%. If the overall importance index of this raw material reaches 9.0, the calculated value is S(9.0) = 85 + 15 * (1 - e^(-1 / 2)) * (87.09 / 2 ... (-0.5) When the overall importance index reaches 10.0, the calculated value is S(10.0) = 85 + 15 * (1 - e) / 90.90%. (-1.0)=94.48%, this exponential growth model enables extremely critical raw materials to obtain near 100% service level assurance.

[0081] For raw materials classified as low importance, a first preset linear function is used for calculation. This function is expressed as S(x) = 75 + 2*x, with a slope of 2. When the overall importance index is 4.2, the calculated target service level percentage is 75 + 2 * 4.2 = 83.4%. For raw materials classified as medium importance, a second preset linear function is used. This function is expressed as S(x) = 70 + 3*x, with a slope of 3. The slope of the second preset linear function is greater than that of the first preset linear function. This slope difference reflects the management strategy that raw materials in the medium importance range experience a faster improvement in service level as the importance index increases. When the overall importance index is 6.5, the calculated target service level percentage is 70 + 3 * 6.5 = 89.5%. This piecewise function design allows for different service level response curves based on the importance level of the raw materials. Low-importance raw materials use a gentle linear growth curve to avoid excessive inventory, medium-importance raw materials use a steeper linear growth curve to balance costs and risks, and high-importance raw materials use an exponential growth curve to ensure the supply security of critical materials.

[0082] After obtaining the target service level percentage, this percentage needs to be converted into a service level coefficient under a standard normal distribution by calling the inverse cumulative distribution function (INCF). The INCF takes the service level percentage as input and returns the Z-value, which represents the cumulative probability from negative infinity to a certain Z-value under the standard normal distribution curve, equal to the input percentage. If the target service level percentage is 87.09%, the function call `norminv(0.8709, 0, 1)` is executed, where 0.8709 is the decimal form of the service level percentage, 0 represents the mean of the standard normal distribution, and 1 represents the standard deviation. The function calculates the Z-value to be approximately 1.13 using a lookup table or numerical iteration method, and outputs this Z-value as the service level coefficient.

[0083] S107: Compare real-time inventory levels with dynamic safety stock thresholds.

[0084] In S107 above, after obtaining the real-time inventory of the target raw material type, the dynamic safety stock threshold of the target raw material type is determined in the above manner, and then the real-time inventory of the target raw material type is compared with the dynamic safety stock threshold to determine whether an early warning is required.

[0085] S108: When the real-time inventory is less than or equal to the dynamic safety stock threshold, a replenishment warning message for the target raw material type is automatically generated and sent to the terminal device of the person in charge of the target through communication, so that the person in charge of the target can carry out the replenishment process.

[0086] In S108 above, when the real-time inventory of the target raw material type is less than or equal to the dynamic safety stock threshold, an early warning for the target raw material type needs to be issued promptly. Detailed attribute information of the target raw material type is extracted from the material information table and combined with the current inventory status to assemble a replenishment warning message. The replenishment warning message includes the target raw material type, current inventory level, dynamic safety stock threshold, inventory shortage quantity, and suggested purchase quantity. Then, the responsible person corresponding to the target raw material type is queried from the responsibility configuration table, and their email address and phone number are retrieved. An email is sent to the responsible person's email address, and warning information is sent via phone and SMS. Upon connection, a pre-recorded voice prompt is played: "Hello, xx inventory is insufficient, currently only x pieces remain, please arrange replenishment as soon as possible." Simultaneously, the buzzer of the on-site silo emits a beeping alarm at a frequency of 2 times per second, along with the continuous flashing of the silo's red indicator light, visually indicating to on-site operators that the silo is in a material shortage state.

[0087] For example, the real-time inventory of zirconia A2 color ceramic blocks is read as 6 blocks, and the dynamic safety stock threshold for zirconia A2 color ceramic blocks is obtained from the inventory threshold table as 20 blocks. When the real-time inventory is less than the dynamic safety stock threshold, a replenishment warning is immediately triggered by the material shortage alarm and notification unit. Detailed attribute information for the target raw material type is extracted from the material information table, including material code, material name, storage location, material color, material size, material transmittance, and material layering information. This information is then combined with the current inventory status data to assemble a replenishment warning information data package. The system retrieves the target responsible person information corresponding to the target raw material type from the responsibility configuration table, including the person's name "Xiao Li," mobile phone number 1385678, email address li@dental-lab.com, and preset phone number. It then uses the integrated GSM communication module to send a text message to the responsible person's mobile terminal: "[Warehouse Warning] Zirconia A2 color, only 6 pieces left in stock, safety threshold 20 pieces, shortage of 14 pieces, it is recommended to purchase 18 pieces immediately, material shortage expected within 2 days." Simultaneously, it sends a replenishment warning email in HTML format to the responsible person's email address via the network module using the SMTP protocol. The email body contains detailed attribute parameters such as the material's color, size, transmittance, and layering information, along with an inventory consumption trend chart and an automatically generated purchase requisition PDF file. Furthermore, it automatically dials the responsible person's preset number 021-5678 via the telephone dialing module. Upon connection, a pre-recorded voice prompt plays: "Hello, the warehousing stock of zirconia A2 color ceramic blocks is insufficient, currently only 6 pieces remain, please arrange replenishment as soon as possible." Simultaneously, the buzzer in the on-site warehousing emits a beeping alarm at a frequency of 2 beeps per second. The multi-channel replenishment early warning mechanism, which integrates SMS, email, telephone, and buzzer, constructs a three-dimensional early warning network. No matter whether the person in charge is in the office, on the production site, or on the road, they can receive replenishment early warning information in a timely manner through at least one communication method. Compared with the traditional method that only relies on on-site sound and light prompts, the multi-channel automatic alarm enables the person in charge to accurately identify the specific raw material specifications that need to be replenished, avoiding procurement errors caused by unclear material information.

[0088] S109: When the real-time inventory level is greater than the dynamic safety stock threshold, confirm that the target raw material type is in a normal state and continue to monitor the inventory level of each raw material type in the warehouse in real time.

[0089] In step S109 above, when the real-time inventory level exceeds the dynamic safety stock threshold, it is confirmed that the target raw material type is currently in a normal state. The presence and sufficient inventory in the silo can be indicated by indicator lights.

[0090] For example, the real-time inventory of zirconia B1 color ceramic blocks is read as 35 blocks, and the dynamic safety stock threshold for this target raw material type is obtained from the inventory threshold table as 22 blocks, confirming that this target raw material type is currently in a normal state. Simultaneously, the status field of this target raw material type is updated to normal in the inventory status table of the central processing system, recording the status update timestamp as 202x-01-22 14:23:15, and this status information is transmitted in real-time to the large screen display via the data push interface. The large screen displays a graphical interface showing a schematic diagram of the silo layout, displaying the raw material type, inventory quantity, and attribute information of each silo stored in each location. Managers can clearly understand through the large screen that the inventory status of this target raw material type fully meets production needs and has sufficient safety margin. Since the target raw material type is determined to be in a normal state, no early warning process from the material shortage alarm and notification unit is triggered. Records related to this target raw material type in previously issued replenishment warnings are marked as "resolved" by the system. If this raw material previously triggered a replenishment warning, a status recovery notification will be automatically sent to the terminal device of the person responsible for the target. After confirming that the target raw material type is in a normal state, the material monitoring and scheduling system continues to monitor the inventory levels of each raw material type in the silo in real time to ensure that all types of raw materials in the silo are in sufficient stock.

[0091] This application also provides an automated silo management system based on real-time monitoring and alarms. The system is a silo management platform, which includes a feeding unit, a detection unit, a processing unit, and an alarm unit. The feeding unit controls an industrial robot based on grasping instructions to grasp raw materials to be stored from the feeding area and place them in the detection area; The detection unit confirms that the raw material to be stored is located in the detection area, then detects the raw material to be stored to obtain raw material attribute information, and determines the raw material type based on the raw material attribute information. The processing unit determines the target storage location among multiple storage locations in the silo based on raw material attribute information and storage location allocation rules. It controls an industrial robot to move the raw material to be stored from the detection area into the target storage location and updates the inventory quantity corresponding to the raw material type in real time based on the receiving action, thus updating the storage status of the target storage location. When the inventory quantity of the target raw material type changes, it retrieves the real-time inventory quantity corresponding to the target raw material type from all storage locations, and obtains historical demand data, procurement cycle data, and average daily consumption corresponding to the target raw material type. The target raw material type can be any raw material type in the storage location. It calculates the demand standard deviation and procurement cycle standard deviation based on the historical demand data and procurement cycle data respectively. It then calculates the dynamic safety stock threshold for the target raw material type based on the average daily consumption, demand standard deviation, and procurement cycle standard deviation. Finally, it compares the real-time inventory quantity with the dynamic safety stock threshold. The alarm unit automatically generates a replenishment warning for the target raw material type when the real-time inventory is less than or equal to the dynamic safety stock threshold. The warning is then sent to the terminal device of the person in charge via communication, so that they can carry out the replenishment process.

[0092] In one possible implementation, the processing unit is used to calculate the average procurement cycle from the procurement cycle data, determine the service level coefficient based on the importance of the target raw material type; multiply the average procurement cycle by the square of the demand standard deviation to obtain the demand fluctuation variance contribution, and multiply the square of the average daily consumption by the square of the procurement cycle standard deviation to obtain the procurement cycle fluctuation variance contribution; add the demand fluctuation variance contribution and the procurement cycle fluctuation variance contribution to obtain the total fluctuation variance, and calculate the square root of the total fluctuation variance to obtain the comprehensive fluctuation standard deviation; multiply the service level coefficient by the comprehensive fluctuation standard deviation to obtain the safety stock; multiply the average daily consumption by the average procurement cycle to obtain the average cycle consumption; and add the average cycle consumption to the safety stock to obtain the dynamic safety stock threshold.

[0093] In one possible implementation, the processing unit is used to acquire multiple importance assessment factors associated with the target raw material type, perform weighted calculations on the multiple importance assessment factors to obtain a comprehensive importance index, the importance assessment factors including clinical criticality level, customization level, and supply scarcity level; compare the comprehensive importance index with a first importance threshold and a second importance threshold, wherein the second importance threshold is greater than the first importance threshold; when the comprehensive importance index is less than the first importance threshold, the target raw material type is classified into a low importance interval, and the target service level percentage is calculated based on a first preset linear function; when the comprehensive importance index is less than the first importance threshold, the target raw material type is classified into a low importance interval, and the target service level percentage is calculated based on a first preset linear function; when the comprehensive importance index is less than the first importance threshold, the target raw material type is classified into a low importance interval, and the target service level percentage is calculated based on a first preset linear function. When the overall importance index is greater than or equal to the first importance threshold and less than the second importance threshold, the target raw material type is classified into the medium importance interval, and the target service level percentage is calculated based on the second preset linear function, the slope of which is greater than the slope of the first preset linear function. When the overall importance index is greater than or equal to the second importance threshold, the target raw material type is classified into the high importance interval, and the target service level percentage is calculated based on the preset exponential function, wherein the target service level percentage increases at an accelerated rate with the increase of the overall importance index. The service level coefficient is obtained by calculating the target service level percentage based on the inverse cumulative distribution function.

[0094] In one possible implementation, the processing unit is used to confirm that the raw material to be stored is located in the target silo, update the real-time storage status associated with the target silo to the occupied status; obtain the unique identifier of the target silo, associate and map the unique identifier with the raw material attribute information, and store the mapping relationship in the silo status database; the indication unit is used to read the real-time storage status of each silo in the silo, if the real-time storage status of the current silo is occupied, control the status indicator light corresponding to the current silo to display a first color; if the real-time storage status of the current silo is idle, control the status indicator light corresponding to the current silo to display a second color, the first color and the second color being different.

[0095] In one possible implementation, the processing unit is used to obtain the raw material type, usage frequency, and chemical compatibility list of the raw material to be stored; obtain the real-time status and access path distance of all silos in the silo, where the real-time status includes occupied or idle status, and the access path distance is the path distance from the starting point of the industrial robot to each silo; filter all silos with an idle real-time status to obtain an initial set of available silos; obtain the adjacent silos of the idle silos from the initial set of available silos, and if the raw material already stored in the adjacent silos is marked as incompatible with the raw material to be stored in the chemical compatibility list, then remove the idle silos from the initial set of available silos to obtain a set of candidate silos, where the idle silos are any silos in the initial set of available silos; and determine the target silo from the set of candidate silos based on usage frequency.

[0096] In one possible implementation, the processing unit is used to divide all storage locations in the silo into high-frequency, medium-frequency, and low-frequency regions based on the access path distance. When the usage frequency is greater than a preset first frequency, storage locations located in the high-frequency region are selected from the candidate storage location set to obtain a high-frequency candidate subset, and the storage location with the smallest access path distance is selected from the high-frequency candidate subset as the target storage location. If there are no candidate storage locations in the high-frequency region, the storage location with the smallest access path distance is selected from the candidate storage locations in the medium-frequency and low-frequency regions as the target storage location. When the usage frequency is less than or equal to the preset first frequency and greater than a preset second frequency, storage locations located in the medium-frequency region are selected from the candidate storage location set. The system first selects a mid-frequency candidate subset, where the first frequency is greater than the second frequency. It then selects the position with the shortest access path distance from the mid-frequency candidate subset as the target position. If no candidate position exists in the mid-frequency region, the system selects the position with the shortest access path distance from the candidate positions in the high-frequency and low-frequency regions as the target position. When the frequency used is less than or equal to the second frequency, the system filters out positions located in the low-frequency region from the candidate position set, obtaining a low-frequency candidate subset. It then selects the position with the longest access path distance from the low-frequency candidate subset as the target position. If no candidate position exists in the low-frequency region, the system selects the position with the longest access path distance from the candidate positions in the mid-frequency and high-frequency regions as the target position.

[0097] In one possible implementation, the detection unit is used to scan the raw material to be stored using a laser scanner to obtain three-dimensional point cloud data, calculate the size information of the raw material to be stored based on the three-dimensional point cloud data, and determine the thickness information based on the size information; acquire side color images and top color images of the raw material to be stored; perform color space analysis on the top color image to obtain color information; segment and extract the side color image to obtain a raw material color image, perform signal conversion on the raw material color image to obtain multiple projection signals; calculate the derivatives of the multiple projection signals to obtain multiple gradients, calculate the peak values ​​from the multiple gradients to obtain peak coordinates, and output the peak coordinates as layering information; measure the light intensity value after light penetrates the raw material to be stored, and obtain the initial light intensity value before the raw material to be stored is placed, and use the ratio of the light intensity value to the initial light intensity value as the transmittance; output the size information, thickness information, color information, layering information, and transmittance as raw material attribute information.

[0098] It should be noted that the system provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0099] This application also discloses an electronic device. (See reference...) Figure 3 , Figure 3 This application provides a schematic diagram of the structure of an electronic device. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 302, and at least one communication bus 305.

[0100] The communication bus 305 is used to enable communication between these components.

[0101] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0102] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0103] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 302, and by calling data stored in memory 302. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and application requests; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0104] The memory 302 may include random access memory (RAM) or read-only memory. Optionally, the memory 302 may include a non-transitory computer-readable storage medium. The memory 302 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 302 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), instructions for implementing the various method embodiments described above, etc. The data storage area may store data involved in the various method embodiments described above. Optionally, the memory 302 may also be at least one storage device located remotely from the aforementioned processor 301.

[0105] like Figure 3 As shown, the memory 302, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application for automated silo management based on real-time monitoring and alarms.

[0106] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 301 can be used to call the application program for automated warehouse management based on real-time monitoring and alarm stored in the memory 302. When executed by one or more processors, the electronic device performs one or more of the methods described in the above embodiments.

[0107] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0108] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0109] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some service interfaces; indirect couplings or communication connections between devices or units may be electrical or other forms.

[0110] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0111] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0112] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0113] The above description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure.

Claims

1. An automated silo management method based on real-time monitoring and alarm, characterized in that, The method, applied to a silo management platform, includes: The industrial robot is controlled by grasping instructions to grasp the raw materials to be stored from the feeding area and place the raw materials to be stored in the detection area; After confirming that the raw material to be stored is located in the detection area, the raw material to be stored is detected to obtain raw material attribute information, and the raw material type is determined based on the raw material attribute information; Based on the raw material attribute information and storage location allocation rules, a target storage location is determined among multiple storage locations in the silo. The industrial robot is controlled to move the raw material to be stored from the detection area into the target storage location. The inventory quantity corresponding to the raw material type is updated in real time based on the warehousing action, and the storage status of the target storage location is updated. When the inventory of the target raw material type changes, the real-time inventory of the target raw material type is obtained from all warehouses, and the historical demand data, procurement cycle data and average daily consumption corresponding to the target raw material type are obtained. The target raw material type is any raw material type in the warehouse. The standard deviation of demand and the standard deviation of procurement cycle are calculated separately from the historical demand data and the procurement cycle data. The dynamic safety stock threshold for the target raw material type is obtained by calculating the average daily consumption, the standard deviation of demand, and the standard deviation of procurement cycle. Compare the real-time inventory level with the dynamic safety stock threshold; When the real-time inventory level is less than or equal to the dynamic safety stock threshold, a replenishment warning message for the target raw material type is automatically generated and sent to the terminal device of the person in charge of the target via communication, so that the person in charge of the target can carry out the replenishment process.

2. The method according to claim 1, characterized in that, The calculation of the average daily consumption, the standard deviation of demand, and the standard deviation of the procurement cycle to obtain the dynamic safety stock threshold for the target raw material type specifically includes: The average procurement cycle is calculated from the procurement cycle data, and the service level coefficient is determined based on the importance of the target raw material type. The contribution of demand fluctuation variance is obtained by multiplying the average procurement cycle by the square of the standard deviation of demand, and the contribution of procurement cycle fluctuation variance is obtained by multiplying the square of the average daily consumption by the square of the standard deviation of the procurement cycle. The total variance is obtained by adding the contribution of the demand fluctuation variance to the contribution of the procurement cycle fluctuation variance, and the square root of the total variance is calculated to obtain the comprehensive standard deviation of the fluctuation. Multiply the service level coefficient by the overall standard deviation of volatility to obtain the safety stock level; Multiply the average daily consumption by the average procurement cycle to obtain the average cycle consumption; The dynamic safety stock threshold is obtained by adding the average periodic consumption to the safety stock level.

3. The method according to claim 2, characterized in that, The determination of the service level coefficient based on the importance of the target raw material type specifically includes: Multiple importance assessment factors associated with the target raw material type are obtained, and a weighted calculation is performed on the multiple importance assessment factors to obtain a comprehensive importance index. The importance assessment factors include clinical criticality level, degree of customization level, and supply scarcity level. The overall importance index is compared with a first importance threshold and a second importance threshold, wherein the second importance threshold is greater than the first importance threshold; When the comprehensive importance index is less than the first importance threshold, the target raw material type is classified into the low importance interval, and the target service level percentage is calculated based on the first preset linear function; When the comprehensive importance index is greater than or equal to the first importance threshold and less than the second importance threshold, the target raw material type is classified into the medium importance interval, and the target service level percentage is calculated based on the second preset linear function, wherein the slope of the second preset linear function is greater than the slope of the first preset linear function. When the comprehensive importance index is greater than or equal to the second importance threshold, the target raw material type is classified into the high importance interval, and the target service level percentage is calculated based on the preset exponential function, wherein the target service level percentage increases rapidly with the increase of the comprehensive importance index. The service level coefficient is obtained by calculating the target service level percentage based on the inverse cumulative distribution function.

4. The method according to claim 1, characterized in that, After updating the inventory quantity corresponding to the raw material type in real time based on the warehousing action and updating the storage status of the target warehouse, the method further includes: After confirming that the raw material to be stored is located in the target storage location, the real-time storage status associated with the target storage location will be updated to an occupied status. Obtain the unique identifier of the target warehouse, associate and map the unique identifier with the raw material attribute information, and store the mapping relationship in the warehouse status database; Read the real-time storage status of each compartment in the silo. If the real-time storage status of the current compartment is occupied, control the status indicator light corresponding to the current compartment to display the first color. If the real-time storage status of the current storage location is idle, the status indicator light corresponding to the current storage location is controlled to display a second color, which is different from the first color.

5. The method according to claim 1, characterized in that, The step of determining the target storage location among multiple storage locations in the silo based on the raw material attribute information and storage location allocation rules specifically includes: Obtain a list of the raw material types, usage frequencies, and chemical compatibility of the raw materials to be stored; The real-time status and access path distance of all bins in the silo are obtained. The real-time status includes occupied status or idle status, and the access path distance is the path distance from the starting point of the industrial robot to each bin. Filter all the positions whose real-time status is idle to obtain an initial set of available positions; Obtain adjacent warehouses of the available warehouses from the initial set of available warehouses. If the raw materials already stored in the adjacent warehouses are marked as incompatible with the raw materials to be stored in the chemical compatibility list, then remove the available warehouses from the initial set of available warehouses to obtain a candidate warehouse set. The available warehouses are any one of the warehouses in the initial set of available warehouses. The target position is determined from the candidate position set based on the frequency of use.

6. The method according to claim 5, characterized in that, The step of determining the target position from the candidate position set based on the usage frequency specifically includes: Based on the access path distance, all storage locations in the silo are divided into high-frequency, medium-frequency, and low-frequency areas. When the usage frequency is greater than a preset first frequency, the warehouse located in the high-frequency region is selected from the candidate warehouse set to obtain a high-frequency candidate subset, and the warehouse with the smallest access path distance is selected from the high-frequency candidate subset as the target warehouse. If there are no candidate positions in the high-frequency region, then the position with the smallest access path distance is selected from the candidate positions in the mid-frequency region and the low-frequency region as the target position. When the usage frequency is less than or equal to the preset first frequency and greater than the preset second frequency, then the positions located in the mid-frequency region are selected from the candidate position set to obtain a mid-frequency candidate subset, wherein the preset first frequency is greater than the preset second frequency; The warehouse with the smallest access path distance is selected from the mid-frequency candidate subset as the target warehouse; If there are no candidate positions in the mid-frequency region, then the position with the smallest access path distance is selected from the candidate positions in the high-frequency region and the low-frequency region as the target position. When the usage frequency is less than or equal to the preset second frequency, the positions located in the low-frequency region are selected from the candidate position set to obtain a low-frequency candidate subset; The warehouse with the largest access path distance is selected from the low-frequency candidate subset as the target warehouse. If there are no candidate positions in the low-frequency region, the position with the largest access path distance is selected from the candidate positions in the mid-frequency region and the high-frequency region as the target position.

7. The method according to claim 1, characterized in that, The step of detecting the raw material to be stored to obtain raw material attribute information specifically includes: The raw material to be stored is scanned using a laser scanner to obtain three-dimensional point cloud data. The size information of the raw material to be stored is obtained by calculating the three-dimensional point cloud data, and the thickness information is determined based on the size information. Acquire side and top color images of the raw material to be stored; Color space analysis is performed on the top surface color image to obtain color information; The side color image is segmented and extracted to obtain a raw material color image, and the raw material color image is signal converted to obtain multiple projection signals; The derivatives of the multiple projection signals are calculated to obtain multiple gradients, and the peak values ​​are calculated from the multiple gradients to obtain peak coordinates. The peak coordinates are then output as layering information. The light intensity value after light passes through the raw material to be stored is measured, and the initial light intensity value before the raw material to be stored is obtained. The ratio of the light intensity value to the initial light intensity value is used as the transmittance. The size information, thickness information, color information, layering information, and light transmittance are output as the raw material attribute information.

8. An automated silo management system based on real-time monitoring and alarm, characterized in that, The system is a silo management platform, which includes a feeding unit, a detection unit, a processing unit, and an alarm unit. The feeding unit controls an industrial robot to grab raw materials to be stored from the feeding area based on grasping instructions, and places the raw materials to be stored in the detection area; The detection unit, after confirming that the raw material to be stored is located in the detection area, detects the raw material to be stored to obtain raw material attribute information, and determines the raw material type based on the raw material attribute information. The processing unit determines a target storage location among multiple storage locations in the silo based on the raw material attribute information and storage location allocation rules. It then controls the industrial robot to move the raw material to be stored from the detection area into the target storage location. Based on the receiving action, it updates the inventory quantity corresponding to the raw material type in real time and updates the storage status of the target storage location. When the inventory quantity of the target raw material type changes, it retrieves the real-time inventory quantity corresponding to the target raw material type from all storage locations, and obtains historical demand data, procurement cycle data, and average daily consumption corresponding to the target raw material type. The target raw material type can be any raw material type in the storage location. It calculates the demand standard deviation and procurement cycle standard deviation based on the historical demand data and procurement cycle data respectively. Finally, it calculates the dynamic safety stock threshold for the target raw material type based on the average daily consumption, the demand standard deviation, and the procurement cycle standard deviation. Compare the real-time inventory level with the dynamic safety stock threshold; When the real-time inventory is less than or equal to the dynamic safety stock threshold, the alarm unit automatically generates a replenishment warning for the target raw material type and sends the replenishment warning to the terminal device of the person in charge of the target through communication, so that the person in charge of the target can carry out the replenishment process.

9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images