Tool Management System and Method Based on Intelligent Tool Cabinet

Abstract

This application provides a tool management system and method based on an intelligent tool cabinet, relating to the field of intelligent tool cabinets. The method includes: a system comprising a data acquisition module, a data processing module, and a tool replenishment plan generation module; the data acquisition module is used to acquire tool inventory data and product order data; the data processing module is used to determine a product production plan based on the product order data; the tool replenishment plan generation module is used to determine a tool usage plan based on the product production plan, and to determine a tool replenishment plan based on the tool usage plan and inventory data. This system improves tool management, increases production efficiency, and reduces inventory costs.

Description

Technical Field

[0001] This application relates to the field of smart vending machines, and more particularly to a knife management system and method based on a smart knife cabinet. Background Technology

[0002] In the machining industry, cutting tools are indispensable consumables, and their management level directly affects production efficiency, product quality, and manufacturing costs. Traditional cutting tool management relies mainly on manual experience and simple information technology, which has many shortcomings. Among them, the most prominent problem is the disconnect between cutting tool inventory management and workshop production planning. The inability to integrate cutting tool management with production leads to a decrease in production efficiency. Summary of the Invention

[0003] This application provides a tool management system and method based on an intelligent tool cabinet, which can improve tool management, increase production efficiency, and reduce inventory costs.

[0004] In a first aspect, this application provides a system for use in an intelligent tool cabinet, the system comprising: a data acquisition module, a data processing module, and a tool replenishment plan generation module;

[0005] The data acquisition module is used to acquire tool inventory data and product order data;

[0006] The data processing module is used to determine the product production plan based on product order data;

[0007] The tool replenishment plan generation module is used to determine the tool usage plan based on the product production plan, and to determine the tool replenishment plan based on the tool usage plan and inventory data.

[0008] In the above technical solution, the data acquisition module automatically collects tool inventory data and product order data, providing an accurate and real-time data foundation for subsequent production planning and tool demand forecasting. Traditional manual inventory and data entry methods are prone to errors and data updates are delayed. However, this system utilizes IoT technology to automate and intelligently collect data, greatly improving the accuracy and timeliness of the data and laying a solid foundation for intelligent decision-making.

[0009] Based on the collected data, the data processing module uses intelligent algorithms to automatically transform order data into structured production plans, clearly defining key elements such as production batches, quantities, and process routes. This automated plan generation method overcomes the inefficiency and errors of manual plan creation, and the generated plans are more executable. The system can also dynamically adjust the production plan according to actual production progress and order changes, making production arrangements more flexible and reasonable, and significantly improving production efficiency and response speed.

[0010] Guided by the production plan, the tooling replenishment plan generation module plays a crucial role. It comprehensively utilizes big data analytics and machine learning technologies to establish a precise tooling demand forecasting model. By analyzing historical tooling consumption data and uncovering the intrinsic correlation between tool wear and production processes and product parameters, it can accurately predict the tooling demand for each process at different time points, forming a demand-driven and precisely prepared tooling plan. Compared to traditional reactive replenishment models, this proactive demand forecasting allows tooling supply to intervene in advance, precisely matching production needs and minimizing downtime caused by tooling shortages. Simultaneously, the tooling replenishment plan generation module fully considers constraints such as inventory status, procurement cycle, and minimum order quantity, using optimization algorithms such as integer programming and heuristic search to automatically generate the optimal tooling replenishment plan. This plan ensures a dynamic balance between tooling supply and demand while minimizing inventory backlog and procurement costs, achieving a perfect combination of economic efficiency and applicability in tooling management.

[0011] A second aspect of this application provides a tool management method based on an intelligent tool cabinet, the method comprising:

[0012] Obtain tool inventory data and product order data, and determine the product production plan based on the product order data;

[0013] Based on the product production plan, determine the tooling plan, and based on the tooling plan and inventory data, determine the tooling replenishment plan.

[0014] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0015] 1. The data acquisition module of this application automatically collects tool inventory data and product order data, providing an accurate and real-time data foundation for subsequent production planning and tool demand forecasting. Traditional manual inventory and data entry methods are prone to errors and data updates are delayed. This system utilizes Internet of Things (IoT) technology to automate and intelligently collect data, greatly improving the accuracy and timeliness of the data and laying a solid foundation for intelligent decision-making.

[0016] 2. Based on the collected data, the data processing module uses intelligent algorithms to automatically transform order data into a structured production plan, clearly defining key elements such as production batches, quantities, and process routes. This automated plan generation method overcomes the inefficiency and errors of manual plan preparation, and the generated plan is more executable. The system can also dynamically adjust the production plan according to actual production progress and order changes, making production arrangements more flexible and reasonable, and significantly improving production efficiency and response speed.

[0017] 3. Under the guidance of the production plan, the tooling replenishment plan generation module plays a crucial role in this application. It comprehensively utilizes big data analytics and machine learning technologies to establish a precise tooling demand forecasting model. By analyzing historical tooling consumption data and uncovering the intrinsic correlation between tool wear and production processes and product parameters, it can accurately predict the tooling demand for each process at different time points, forming a demand-driven and precisely prepared tooling plan. Compared to the traditional reactive replenishment model, this forward-looking demand forecasting allows tooling supply to intervene earlier, precisely matching production needs and minimizing downtime caused by tooling shortages. Attached Figure Description

[0018] Figure 1 An architecture diagram of a tool management system based on an intelligent tool cabinet is provided for embodiments of this application;

[0019] Figure 2 An architecture diagram of a tool replenishment plan generation module provided in this application;

[0020] Figure 3 This is a flowchart illustrating a tool management method based on an intelligent tool cabinet, provided as an embodiment of this application. Detailed Implementation

[0021] 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.

[0022] 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.

[0023] 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 with "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.

[0024] To facilitate understanding of the methods and systems provided in the embodiments of this application, the background of the embodiments of this application will be introduced before introducing the embodiments of this application.

[0025] In modern manufacturing, cutting tools are indispensable consumables in machining, and their management level directly affects production efficiency, product quality, and manufacturing costs. Traditional cutting tool management mainly relies on manual experience and simple information technology, which has many shortcomings. Among them, the most prominent problem is the lack of intelligence and precision in cutting tool replenishment decisions, making it difficult to adapt to the requirements of the era of intelligent manufacturing.

[0026] Specifically, traditional tool replenishment decisions are often based on fixed rules of thumb or simple statistical models, lacking comprehensive consideration and dynamic optimization of key factors such as production planning, tool consumption, and inventory status. Tool management departments lack accurate and timely understanding of real-time tool demand in the workshop, and their analysis of the consumption patterns and influencing factors of various tools is insufficient, making it difficult to predict and accurately prepare inventory. Furthermore, most existing replenishment strategies adopt a reactive, reactive approach, triggering replenishment only when tool inventory drops to minimum levels. This delayed replenishment response often leads to tool shortages on the production floor, causing production stoppages or decreased product quality.

[0027] Furthermore, when production plans or process parameters change, existing replenishment plans are difficult to adjust and optimize quickly, and the types, quantities, and timing of replenishment are hard to match precisely with new production needs. When large quantities of special-specification cutting tools are required, the purchasing department often struggles to complete procurement in a short time, impacting production schedules. Simultaneously, due to a lack of refined demand forecasting and differentiated inventory management, a "polarization" of cutting tool inventory often occurs: some tools are heavily stockpiled, tying up capital and inventory space, while others are frequently out of stock, failing to meet production needs.

[0028] After the background introduction above, those skilled in the art can understand the problems existing in the prior art. The technical solutions in the embodiments of this application 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.

[0029] Based on the aforementioned background technology, further please refer to... Figure 1 , Figure 1 This application provides an architecture diagram of a tool management system based on an intelligent tool cabinet. The system can be implemented using a computer program or run as an independent tool application. Specifically, in this application embodiment, the system can be applied to a server, but it can also be applied to electronic devices such as servers. The system is applied to an intelligent tool cabinet and includes: a data acquisition module 1, a data processing module 2, and a tool replenishment plan generation module 3.

[0030] The data acquisition module is used to acquire tool inventory data and product order data;

[0031] Specifically, in this embodiment, the data acquisition module fully utilizes technologies such as the Internet of Things (IoT), RFID, and sensors to achieve automated and intelligent data collection. Specifically, the intelligent knife cabinet is equipped with an RFID reader and electronic tags, with each knife bearing a unique electronic tag. When a knife enters or leaves the intelligent knife cabinet, the RFID reader automatically identifies the electronic tag, records information such as the knife's type, serial number, borrowing / return time, etc., and updates the knife inventory data in real time. Simultaneously, the data acquisition module also interfaces with the enterprise's ERP, MES, and other systems to automatically acquire product order data, including order number, product model, order quantity, delivery date, and other information.

[0032] Through this automated and intelligent data acquisition method, the data acquisition module can continuously obtain accurate and real-time tool inventory data and product order data, providing reliable data support for subsequent production planning and tool demand forecasting. On the one hand, the system grasps the real-time status and full lifecycle data of each tool, enabling refined management of tool use, maintenance, and disposal, improving tool utilization and extending service life. On the other hand, the system senses changes in market demand in real time, allowing for timely adjustments to production plans and tool allocation, rapid response to order demands, and improved delivery capabilities and customer satisfaction.

[0033] The data processing module is used to determine the product production plan based on product order data;

[0034] Specifically, the data processing module fully utilizes intelligent algorithms and big data analytics to achieve automatic generation and dynamic optimization of production plans. Specifically, after acquiring product order data in real time from the data acquisition module, the data processing module first preprocesses the order data, converting it into a standardized data format and classifying and sorting it according to attributes such as delivery date and product model. Then, the system calls production plan optimization algorithms, such as heuristic search algorithms, genetic algorithms, and ant colony algorithms, to automatically generate the optimal production plan based on multiple factors including production resources, process constraints, and delivery dates.

[0035] The generated product production plans are presented intuitively in the form of spreadsheets or Gantt charts, clearly listing key elements such as production batches, quantities, required materials, process routes, critical procedures, and completion times for each order, providing precise guidance for workshop production. Compared with manual planning, this intelligent and automated planning generation method is not only more efficient, but also significantly improves the accuracy and executability of the plans.

[0036] Meanwhile, the data processing module also features dynamic production plan optimization. In actual production, unexpected situations such as equipment failures, material shortages, and order changes often arise, making it difficult to execute the original plan. The data processing module, by acquiring real-time production execution data and continuously monitoring plan execution, automatically re-optimizes the production plan upon detecting deviations, adjusting the priority and sequence of production tasks to ensure delivery requirements are met to the greatest extent possible within limited resource constraints. This dynamic, real-time plan optimization keeps the production plan in an optimal state, significantly improving the company's ability to respond to changes.

[0037] The tool replenishment plan generation module is used to determine the tool usage plan based on the product production plan, and to determine the tool replenishment plan based on the tool usage plan and inventory data.

[0038] Specifically, the replenishment plan generation module first automatically determines the corresponding tooling plan based on the product production plan generated by the data processing module. The system deeply analyzes each process in the production plan and, in conjunction with factors such as process parameters, processing materials, and equipment type, uses intelligent algorithms to accurately predict the tooling consumption of each process at different time periods. This bottom-up demand forecasting method overcomes the limitations of traditional rules of thumb and simple statistical models, fully considering the dynamic changes and complex constraints of the production process, thus significantly improving the accuracy of tooling demand forecasting.

[0039] After generating a precise tooling plan, the tooling replenishment plan generation module combines real-time tooling inventory data with factors such as procurement lead time, supplier response time, transportation cycle, and budget. Through intelligent optimization algorithms, it automatically generates the optimal tooling replenishment plan. This plan specifies the procurement category, quantity, timeframe, and supplier selection for each type of tool, striving to minimize inventory and procurement costs while ensuring tooling supply.

[0040] Based on the above embodiments, as an optional embodiment, please refer to... Figure 2 The tool replenishment plan generation module 3 includes a production plan decomposition module 31, a production process analysis module 32, a tool consumption prediction module 33, and a tool usage plan generation module 34, which includes:

[0041] The production plan decomposition module is used to decompose the production plan to obtain production batches, production quantities, and production process flows.

[0042] Specifically, after receiving the product production plan generated by the data processing module, the production planning decomposition module first classifies the plan according to product model and delivery date, forming several independent production batches. Each production batch corresponds to a certain number of similar products and has a relatively concentrated production time window. This batch-based planning decomposition can avoid mutual interference between different products and orders, and is conducive to improving the orderliness and targeting of production organization.

[0043] Next, the production planning decomposition module further breaks down each production batch, determining the specific production quantity and corresponding production process. The production quantity directly determines the processing workload of that batch in each process step, and is a key factor in matching tooling requirements. The production process describes key information such as each process step, processing parameters, equipment requirements, and quality standards throughout the entire product manufacturing process, and is an important basis for precise tooling allocation.

[0044] During the decomposition process, the production planning decomposition module fully utilizes information resources such as the process knowledge base, product BOM, and manufacturing execution system to quickly identify the characteristics of production batches, match them with standard process flows, and generate a structured and parameterized production task list. This intelligent decomposition method not only ensures the accuracy and standardization of the decomposition results but also greatly improves decomposition efficiency and reduces the burden of manual processing.

[0045] Through the production planning decomposition module, the originally general production plan is transformed into a series of specific production batches, each with a clear production quantity and process flow, providing a precise task carrier for subsequent tooling demand forecasting. This production process-oriented planning decomposition allows tooling allocation decisions to be closer to actual production, reducing distortion of demand information during transmission and thus improving the matching degree between tooling supply and production needs.

[0046] Meanwhile, standardized and structured production process information provides a data foundation for intelligent algorithms to analyze tool consumption patterns and build predictive models. The system can fully explore the correlations between different batches and processes, identify key factors affecting tool demand, and form a more universal and accurate predictive model. This data-driven intelligent prediction, combined with refined production plan decomposition, can achieve tool demand prediction at the "day" and "piece" granularity, making tool allocation more precise and efficient.

[0047] The production process analysis module is used to determine the type and quantity of cutting tools required for each process based on the production process flow.

[0048] Specifically, after receiving the production batch and process flow information output by the production plan decomposition module, the production process analysis module performs in-depth analysis of the process flow for each batch. By comparing product drawings, process documents, and basic process data, the system automatically identifies the processing characteristics of each process, such as the machined surface, precision requirements, material properties, and batch quantity, and matches the required tool types and quantities accordingly.

[0049] This analytical process typically employs a combination of rule-based expert systems and case-based reasoning systems. First, the system uses a tool selection rule base summarized by domain experts to match machining elements with tool attributes, initially determining the types of candidate tools. Then, the system searches through a large amount of historical production data for successful cases similar to the current process, comparing and analyzing data such as the types, quantities, and consumption of tools used to optimize and refine the initial selection, ultimately generating a targeted tooling list.

[0050] Through the processing module of the production process flow analysis, the system can accurately and efficiently calculate the tooling requirements for each stage of production for a specific product or batch, ensuring a precise mapping between tooling types and quantities and machining tasks. Compared to the traditional method of relying on experience and technician memory to determine tooling, this intelligent and data-driven process analysis can reduce errors and omissions caused by human factors by more than 90%, laying a reliable foundation for precise tooling allocation. Simultaneously, the data on the correspondence between process characteristics and tooling selection accumulated during the analysis process creates conditions for subsequent knowledge accumulation and optimized application. By continuously enriching and improving the process-tooling knowledge base, the system can predict tooling requirements, assess procurement costs, and guide design optimization during the process design stage. Designers can also fully utilize historical data, benchmark against best practices, and rationally formulate machining parameters and tooling schemes, controlling tooling consumption from the source and improving production efficiency.

[0051] The tool consumption prediction module is used to acquire historical tool consumption data and determine the tool consumption of each process based on the historical tool consumption data and the type and quantity of tools required for each process.

[0052] Specifically, the tool consumption prediction module first connects to the database to extract historical tool consumption data related to the current production task. This data typically includes multi-dimensional information such as tool number, name, supplier, usage, usage duration, corresponding process, product model, and material batch. The system cleans, aligns, and correlates this data to form a structured tool consumption sample set.

[0053] Next, the tool consumption prediction module matches and correlates the types and quantities of tools required for each process step output by the production process analysis module with historical consumption data. The system uses intelligent algorithms to analyze the consumption patterns of each tool under different process parameters and constructs a usage prediction model. Commonly used modeling methods include multiple regression, time series analysis, and neural networks.

[0054] During the modeling process, the system fully considers various factors affecting tool consumption, such as cutting parameters, material properties, equipment characteristics, process parameters, and operating time, striving to comprehensively reveal the inherent laws of tool consumption. Simultaneously, the system comprehensively evaluates the goodness of fit and generalization ability of different prediction models, and through cross-validation and parameter tuning, selects the model with the best performance as the final prediction tool.

[0055] Based on a mature prediction model, the tool consumption prediction module can provide quantitative consumption prediction values ​​for each process and each tool category. This prediction is usually an interval value, representing the confidence interval of tool consumption. It takes into account both the trend of historical data and the volatility of the production process, ensuring the scientific nature and reliability of the prediction results.

[0056] The tool usage plan generation module is used to determine the tool usage plan based on the tool consumption, production batch, and production quantity for each process.

[0057] Specifically, the tool usage plan generation module receives tool consumption data for each process from the tool consumption prediction module, as well as production batch and quantity information from the production plan decomposition module. The system integrates and matches this data to calculate the tool usage for each batch at each process stage, forming a tiered tool usage plan table.

[0058] During the calculation process, the tool usage plan generation module employs a rolling planning method, using production batches as the smallest unit. Combining production quantity and process cycle time, it predicts tool usage intensity for each time period, allocating tool consumption to each batch, each process, each piece of equipment, and each shift, thus creating refined tool usage guidelines. Simultaneously, the system fully considers constraints such as equipment capacity, material supply, and quality requirements, using integer programming and heuristic algorithms to dynamically optimize the tool usage plan, striving to improve tool utilization and shorten setup time while meeting production demands.

[0059] Through the calculations of the tool usage plan generation module, a practical and dynamically optimized tool usage plan is formed. This plan clearly specifies the type, quantity, replacement time, and responsible person for each tool used in each process, detailing each tool while covering the entire product production cycle. Workshop managers can use this plan to prepare the necessary tools, rationally schedule their use, and promptly replace worn tools to ensure continuous production.

[0060] Based on the above embodiments, as an optional embodiment, the tool replenishment plan generation module includes: a tool supply and demand difference determination module and a tool replenishment plan module.

[0061] The tool supply and demand difference determination module is used to compare the tool demand data at each time point in the tool usage plan with the inventory data, calculate the tool supply and demand difference at each time point, and determine the tool supply and demand difference based on the tool supply and demand difference at each time point.

[0062] Specifically, the tool supply-demand difference determination module first obtains the tool usage plan data output by the tool usage plan generation module, extracting the tool demand quantity information at each time point. These time points are usually divided according to the production cycle, including both the current moment and key time nodes in the future, such as shift handover and model switching. The tool demand quantity at each time point reflects the tool usage in the production stage before and after that moment, serving as the benchmark value for inventory management. Next, the tool supply-demand difference determination module reads real-time inventory data from the database to obtain information such as the current inventory quantity and storage location of various types of tools. The system aligns this actual inventory data with the demand data in the tool usage plan in terms of time points and categories, calculating the supply-demand difference for each type of tool at each time point.

[0063] During the calculation process, the system fully considers the impact of factors such as in-transit inventory, safety stock, and procurement lead time. It utilizes methods such as demand forecasting, inventory analysis, and material planning to dynamically assess the supply-demand balance from multiple dimensions. For times when supply exceeds demand, the system issues a surplus warning, prompting managers to promptly reduce inventory and avoid overstocking. For times when supply falls short of demand, the system issues a shortage warning, indicating the categories and quantities of goods requiring replenishment, prompting managers to procure in advance to prevent material shortages. Simultaneously, the system tracks inventory change trends, assesses the severity and scope of supply-demand discrepancies, and reminds managers to pay attention to key periods and critical product categories.

[0064] The tool replenishment planning module is used to determine the tool replenishment plan based on the tool supply and demand difference at each point in time, the preset supplier delivery cycle, and the product production plan.

[0065] Specifically, the tooling replenishment planning module receives tooling supply-demand difference data at each time point from the tooling supply-demand difference determination module, obtains preset supplier delivery cycle parameters, and reads product production plan information for a future period. The system integrates and correlates this data, uses intelligent algorithms to optimize calculations, and forms a tiered replenishment plan table with a time axis as the baseline and product categories as units.

[0066] During the replenishment plan generation process, the system fully utilizes supply and demand gap data to quantitatively assess the risk of material shortages at each point in time. For points with large shortages and tight deadlines, the system automatically matches suppliers with strong supply capabilities and short delivery cycles, prioritizing them for replenishment. For points with small shortages and ample time, the system prioritizes suppliers with lower prices and smaller minimum order quantities, ensuring supply while reducing procurement costs. Simultaneously, the system also considers product production plans, analyzing future peak and off-peak periods for tool usage to predict replenishment schedules in advance. For product categories with upcoming peak usage periods, the system appropriately increases replenishment frequency and quantity to ensure sufficient inventory reserves; for product categories with upcoming off-peak usage periods, the system appropriately reduces replenishment frequency and quantity to minimize inventory buildup. Furthermore, when formulating replenishment plans, the system also reserves a certain amount of safety stock and buffer time to cope with unexpected situations such as supplier defaults, logistics delays, and temporary additional orders, enhancing supply chain resilience.

[0067] Based on the above embodiments, as an optional embodiment, the system further includes: an image acquisition module, a wear degree recognition module, and a tool return and storage module;

[0068] The image acquisition module is used to acquire images of the returned tools.

[0069] Specifically, the return area is equipped with imaging devices such as high-resolution industrial cameras, parallel light sources, and background panels. The camera uses autofocus technology, which can adaptively adjust the focal length and aperture according to the height of the tool to capture tool images with distinct layers and clear textures. The shooting process uses multi-angle, multi-spectral imaging methods. By adjusting the camera position and the angle of the light source, it can obtain the surface detail features of different parts of the tool and different materials, comprehensively reflecting the tool's geometric appearance and material properties.

[0070] The wear degree recognition module is used to determine the wear degree of the returned tool based on the image of the tool;

[0071] Specifically, the wear level identification module employs transfer learning, introducing attention mechanisms and multi-scale analysis on top of classic CNN network architectures (such as AlexNet, VGGNet, ResNet, etc.) to specifically extract salient features of tool wear. During offline training, a large number of labeled tool wear images are used as learning samples. The gradient backpropagation algorithm is used to continuously optimize model parameters and uncover deeper patterns in different wear modes. In online application, the wear level identification module reads the tool image to be detected uploaded by the image acquisition module, uses the trained CNN model for forward inference calculation, and obtains quantitative scores for the tool on typical wear modes such as edge wear, chipping, cracks, and deformation. These individual scores are then integrated to form a comprehensive wear evaluation level (such as "intact," "lightly worn," "severely worn," "scrapped," etc.). The entire process is completed automatically without manual intervention.

[0072] Through intelligent identification by the wear level recognition module, an objective and quantitative wear detection report is generated every time a tool is returned. Compared to traditional methods relying on human observation and subjective judgment, this method can improve the accuracy of wear assessment by more than 10 times and increase detection efficiency by more than 100 times. Managers can use the wear level data to accurately grasp the health status and remaining life of each tool, providing a reliable basis for subsequent tool maintenance and usage decisions.

[0073] Based on the above embodiments, as an optional embodiment, the wear degree identification module includes: a feature extraction module and a wear identification module;

[0074] The feature extraction module is used to extract key features through image segmentation algorithms;

[0075] Specifically, the feature extraction module employs a deep learning-based semantic segmentation method, separating the wear region from the background through hierarchical feature extraction and pixel-level annotation. During offline training, a large number of labeled wear region sample images are used as ground truth labels. Classic semantic segmentation networks such as fully convolutional networks and U-shaped networks are used for end-to-end learning, enabling the system to master high-level semantic features such as the shape, contour, and boundary of the wear region. In online applications, the feature extraction module inputs the image of the tool to be detected into the trained segmentation network, automatically inferring the category (wear / non-wear) of each pixel in the image and generating a complete wear region segmentation mask.

[0076] After obtaining the mask of the worn area, the feature extraction module further utilizes image processing algorithms to extract the salient features of the region, including geometric features such as area, perimeter, major axis, and minor axis; texture features such as grayscale mean, contrast, and entropy; and key point features such as scale-invariant feature transformation and acceleration robustness features, forming a complete feature vector that characterizes the morphological attributes of the worn area. To further eliminate differences in image scale and rotation, the system also performs preprocessing such as normalization and whitening on the features to improve their robustness.

[0077] Through the refined processing of the feature extraction module, complex tool wear images are transformed into standardized wear area feature matrices, and the detailed features of the wear are quantified into calculable numerical indicators.

[0078] The wear identification module is used to obtain the wear degree of the returned tool by inputting a preset wear degree assessment model based on key features.

[0079] Specifically, the wear identification module employs a pre-defined wear level assessment model, which is built using common machine learning classification algorithms such as Support Vector Machines and Random Forests. During the offline training phase, the wear area feature vectors output by the feature extraction module and manually labeled wear level pairs are used to form training sample pairs. These pairs are then input into the classification model for parameter learning and hyperparameter tuning, continuously fitting and optimizing the discrimination boundary between features and labels. Through strategies such as cross-validation and grid search, the model achieves optimal classification accuracy on both the training and validation sets.

[0080] In online applications, the wear identification module reads the wear features of the tool to be identified from the feature extraction module. Based on these key features, it inputs them into a pre-trained wear assessment model to automatically predict the wear level label of the tool (such as "light wear," "moderate wear," "severe wear," etc.). Simultaneously, the model provides a confidence probability for the classification result, reflecting the reliability of the decision. When the confidence level is too low, the system automatically triggers a manual review process, improving the accuracy of the classification through human-machine collaboration.

[0081] Through the wear identification module, the degree of tool wear is quantified into a clear health level. Compared with manual assessment, this method can reduce the error rate by more than 80% and improve the assessment efficiency by more than 100 times. Quantifying wear into level labels also facilitates visualization and statistical analysis, allowing managers to intuitively understand the overall wear distribution of tool groups and optimize production organization. Simultaneously, wear level data can also be used to predict the remaining service life of tools, enabling predictive maintenance.

[0082] The tool return and inventory module is used to classify returned tools according to their wear level, obtain category data of returned tools, and generate inventory tool category data based on the category data of returned tools and the tool inventory data.

[0083] Specifically, the purpose of setting up the returned tool storage module is to refine the management of tools in different health states and improve inventory turnover efficiency. Traditional tool management often adopts a "one-size-fits-all" approach, treating all returned tools the same, without distinguishing between good and bad, resulting in problems such as large performance differences and mismatched uses, which affects production efficiency.

[0084] In this embodiment, returned tools are categorized into "intact," "repairable," and "scrap" based on their wear level. Intact tools are prioritized for use, repairable tools are repaired centrally, and scrapped tools are replaced promptly, maximizing the remaining value of the tools. Furthermore, based on production task characteristics, the system further subdivides "intact" tools into subcategories such as "finishing tools" and "roughing tools," achieving precise matching between tool performance and process requirements.

[0085] In practice, the tool return module reads the tool wear level data output by the wear level identification module and automatically determines the tool's category label based on preset classification rules (such as wear score thresholds). Then, the system controls an AGV (Automated Guided Vehicle) to transport the tool to the corresponding storage location. The storage location is equipped with an electronic tag that senses the number of tools entering the storage. Simultaneously, the system collects real-time environmental parameters (such as temperature and humidity) of the storage location to optimize storage conditions.

[0086] Meanwhile, the tool return and warehousing module also updates the tool inventory records in the database. The system matches and merges the category data of returned tools with the original inventory data to generate the latest inventory tool category data. This data includes dimensions such as the quantity, percentage, and age of each category ("intact," "under repair," and "scrap"), comprehensively reflecting the inventory structure. Data is updated in real time to ensure consistency between records and actual inventory. Managers can access this data at any time to optimize inventory strategies. Through the tool return and warehousing module, integrated closed-loop management of tool classification, intelligent storage, and precise allocation can be achieved.

[0087] Based on the above embodiments, as an optional embodiment, the system further includes: an allocation level determination module and a tool allocation module;

[0088] The allocation level determination module is used to obtain the worker's identity information when the worker receives the cutting tools, determine the process flow of the product the worker is responsible for based on the worker's identity information, and determine the allocation level of the cutting tools based on the process flow of the product the worker is responsible for and the product production plan.

[0089] Specifically, the purpose of setting up the allocation level determination module is to allocate tools of matching quality to different production stages and processing needs, thereby improving production efficiency and product quality. Traditional tool requisition often adopts a "first-come, first-served" model, which does not fully consider differences in production progress and process difficulty, resulting in low tool utilization efficiency. This embodiment, however, starts from the entire product production process and establishes a dynamic scheduling mechanism with allocation levels as the bridge, which continuously optimizes tool utilization while meeting production needs.

[0090] In practice, the allocation level determination module employs IoT sensing technologies such as RFID and facial recognition to automatically acquire workers' identity information when they collect tools. The system accesses the workshop production management system, indexes the production task sheet of the product the worker is currently responsible for based on the worker's ID, and analyzes the product's processing steps and progress information. Simultaneously, the system links product design drawings, process specifications, and other data to obtain processing requirements parameters such as precision requirements and surface roughness for each process, forming a comprehensive process requirement profile. This profile is then synchronized in real time with upstream data such as the enterprise's production planning system and material control system, achieving dynamic matching between process requirements and production progress.

[0091] After understanding the product production schedule and process requirements, the allocation level determination module automatically determines the allocation level of the tools requisitioned by the worker using preset allocation strategy rules. The allocation levels are divided into multiple grades from high to low, including "finishing grade," "normal grade," and "roughing grade," with each grade corresponding to different quality levels of tool priority allocation. The rules for determining the allocation level are based on the analysis of a large amount of historical production data, comprehensively considering multiple factors such as the matching degree between production stage and batch size, the matching degree between finishing processes and tool performance, and the urgency of delivery deadlines, forming a multi-objective optimized dynamic allocation model. By subdividing product requirements into different levels, refined matching of tool resources can be achieved, maximizing the utilization of limited high-end tools while ensuring machining quality.

[0092] The tool allocation module allocates tools based on the tool allocation level and inventory tool category data.

[0093] Specifically, the tool allocation module reads the inventory tool data from the warehouse management system in real time, obtaining multi-dimensional attribute information such as the quality grade, quantity, location, and purchase time of each type of tool. Simultaneously, the tool allocation module receives tool allocation level requests from the allocation level determination module, clarifying the required tool quality level for this requisition. The system refines the allocation level into multiple matching rules, including quality, precision, material, and tool type, and retrieves a subset of candidate tools from the inventory tool data that meet these rules.

[0094] Building upon this foundation, the tool allocation module employs heuristic intelligent algorithms, such as genetic algorithms and ant colony algorithms, to search for the optimal allocation combination within a subset of candidate tools. The algorithm aims to "improve tool utilization and reduce inventory backlog," employing allocation strategies that prioritize matching tools with high turnover rates and consuming tools nearing their expiration date. It also considers on-site factors such as the concentration of tool placement locations and the movement distance of workers receiving the tools. Through multiple iterations, it identifies the allocation scheme with the lowest cost and highest efficiency.

[0095] Based on the above embodiments, as an optional embodiment, the allocation level determination module includes: a process flow analysis module, a priority determination module, a tool requirement level determination module, and a tool allocation level determination module;

[0096] The process flow analysis module is used to perform a detailed analysis of the process flow of the products under the worker's responsibility, and to obtain the processing characteristics, quality requirements, process parameters and key processes of each process.

[0097] Specifically, the process flow analysis module first extracts the product's process route data, including the name, processing content, and sequential relationship of each process step. Next, the system performs semantic analysis on the process text, identifying the processing type (e.g., milling, drilling, turning), processing features (e.g., planes, arcs, threads), processing location (e.g., slots, holes, surfaces), and processing accuracy (e.g., dimensional tolerances, geometric tolerances, surface roughness) of each process step. Then, the system retrieves the product's 3D model data and uses virtual simulation and finite element analysis to simulate the stress state of each process step, calculate process parameters (e.g., cutting speed, feed rate, depth of cut), and determine the process difficulty and quality risks. Finally, considering factors such as processing difficulty, quality impact, and production capacity load, the system identifies the key processes in the process route and forms a process knowledge graph based on a semantic network.

[0098] The priority determination module is used to determine the priority of the process flow that workers are responsible for based on the product production plan;

[0099] Specifically, the tool requirement level determination module first extracts the semantic attributes of each process node in the process knowledge graph, focusing on analyzing key information closely related to tool selection, such as machining characteristics, quality requirements, and process parameters. For example, regarding machining characteristics, the system focuses on attributes such as workpiece material, machining accuracy, and surface quality; regarding quality requirements, the system focuses on indicators such as dimensional tolerances, geometric tolerances, and surface roughness; regarding process parameters, the system focuses on values ​​such as cutting speed, depth of cut, and cutting depth. The system uses intelligent algorithms such as ontology reasoning and rule engines to map various attribute values ​​to specific tool performance requirements, such as tool material, cutting edge shape, tool tip radius, and tool holder type. Based on this, the system comprehensively considers factors such as the importance of the process, machining difficulty, quality risk, and production capacity load, classifying all processes into four requirement levels: important, critical, ordinary, and minor. Among them, critical processes and high-risk processes are classified as important, imposing stringent requirements on tool performance and quality; ordinary processes that are simple and easy to machine are classified as minor, with relatively relaxed requirements on tools. The system uses the analytic hierarchy process (AHP) to determine the weights of various factors, forming a quantitative evaluation matrix of process requirement levels and an intelligent decision-making mechanism.

[0100] Next, the tooling demand level determination module automatically matches existing tools in the enterprise's tooling resource pool based on the demand level. For important and critical processes, the system prioritizes high-end tooling products with excellent performance and stable quality, such as alloy tools and ceramic tools, and sets strict usage specifications and discard standards to ensure processing results and product quality. For ordinary and minor processes, the system matches cost-effective and versatile economical tools, meeting basic usage requirements while reducing procurement costs and inventory levels. Simultaneously, the system also evaluates the overall cost of tooling use, including tool price, lifespan, and regrinding frequency, optimizing the tooling lifecycle management strategy. For example, by comparing the cost-effectiveness of different brands and models of tools, the system automatically eliminates tooling models with high usage costs and poor performance; for frequently used general-purpose tools, the system optimizes the economic order quantity through big data analysis to minimize procurement and inventory costs.

[0101] The tooling requirement level determination module is used to determine the tooling requirement level based on the machining characteristics, quality requirements, process parameters, and key processes of each operation.

[0102] Specifically, the tooling requirement level determination module first extracts the semantic attributes of each process node in the process knowledge graph, focusing on analyzing key information closely related to tool selection, such as machining characteristics, quality requirements, and process parameters. For example, regarding machining characteristics, the system focuses on attributes such as workpiece material, machining accuracy, and surface quality; regarding quality requirements, the system focuses on indicators such as dimensional tolerances, geometric tolerances, and surface roughness; regarding process parameters, the system focuses on values ​​such as cutting speed, depth of cut, and cutting depth. The system uses intelligent algorithms such as ontology reasoning and rule engines to map various attribute values ​​to specific tool performance requirements, such as tool material, cutting edge shape, tool tip radius, and tool holder type. Based on this, the system comprehensively considers factors such as the importance of the process, machining difficulty, quality risk, and production capacity load, dividing all processes into four requirement levels: critical, important, ordinary, and auxiliary. Among them, critical-level requirements correspond to key product features and high-risk processes, imposing stringent requirements on tool performance and quality; ordinary processes, which are simple and easy to machine, correspond to auxiliary-level requirements, with relatively relaxed requirements on tools. The system uses the analytic hierarchy process (AHP) to determine the weights of various factors, forming a quantitative evaluation matrix of process requirement levels and an intelligent decision-making mechanism.

[0103] Next, the tooling requirement level determination module automatically matches existing tools from the enterprise's tooling resource pool based on the requirement level. For critical and important processes, the system prioritizes matching high-end tooling products with excellent performance and stable quality, such as CNC tools and high-performance coated tools, and sets strict usage specifications and scrap standards to ensure machining results and product quality. For ordinary and auxiliary processes, the system matches cost-effective and versatile economical tools, meeting basic usage requirements while reducing procurement costs and inventory holdings.

[0104] The tool allocation level determination module is used to determine the tool allocation level based on priority and tool requirement level.

[0105] Specifically, the purpose of determining the tool allocation level is to coordinate production plans and tool resources, scientifically formulate tool usage plans, and avoid over-optimization and resource waste in local processes. Traditional tool management often adopts a "first-come, first-served" allocation principle, lacking overall coordination and dynamic balance, which easily leads to uneven tool allocation and blind grabbing, affecting production progress and product quality. This embodiment, based on a whole-process perspective, takes into account the priority of each process and rationally determines the tool allocation plan. While ensuring key processes, it improves the turnover utilization rate of tool resources, which is a key measure to implement lean management and balance production rhythm.

[0106] In practical implementation, the tool allocation level determination module first extracts production plan and order information to analyze the priority level of processes. The system focuses on factors such as order delivery date, product batch size, and processing difficulty, and uses intelligent algorithms such as multi-criteria decision-making and fuzzy comprehensive evaluation to divide all processes into four priority levels: urgent, priority, normal, and surplus. Among them, sample trial production and urgent revision orders with tight delivery dates correspond to the urgent level, requiring priority to ensure tool supply; normal orders with high difficulty and large batch size correspond to priority, and are given appropriate preferential treatment in tool usage; spare parts with ample delivery dates and low technical difficulty correspond to the surplus level, and are in a lower position in tool allocation. The system uses the analytic hierarchy process (AHP) to determine the weight of various factors and form a quantitative judgment matrix of process priority. Next, the system combines the tool demand level generated by the previous module to combine and map the priority and demand level to form several tool allocation levels. For example, combining the urgent priority with the critical tool demand level corresponds to the highest level of tool allocation, which has the highest authority to use tools; combining the surplus priority with the auxiliary tool demand level corresponds to the lowest level of tool allocation, which is strictly restricted in tool usage. The system matches existing tooling resources to a company in descending order of allocation level. For the highest allocation level, the system will unconditionally meet its tooling needs, prioritizing the allocation of suitable tools from the tooling resource pool. If existing resources are insufficient, an emergency procurement process will be initiated to expedite tooling supply from suppliers. For the lowest allocation level, the system allocates surplus tooling resources based on inventory and usage status, without initiating new tooling purchases. Simultaneously, the system maintains overall balance and adjusts allocation strategies as needed. For example, when there is a severe shortage of tools at a certain allocation level, the system will temporarily lower the allocation level of some processes, temporarily relinquishing tooling access to more urgent orders. Furthermore, once a batch of orders is completed, the system will promptly reallocate its remaining tooling resources to other pending orders, improving tooling utilization efficiency.

[0107] Based on the above embodiments, as an optional embodiment, the allocation level determination module further includes: a worker information acquisition module;

[0108] The worker information acquisition module is used to acquire the worker's facial image when the worker receives the knife, and compare the facial image with a preset image database to obtain the worker's identity information.

[0109] Specifically, the worker information acquisition module is equipped with high-definition cameras, facial recognition software, and other hardware and algorithm models to monitor the tool collection process in real time. When a worker arrives at the tool collection point and requests tools, the system immediately captures a facial image of the worker using the camera, extracting key feature points such as the coordinates and proportions of the eyes, nose, and mouth to create a unique facial feature vector for that worker. The system then compares this feature vector with a pre-set facial image database. This database stores high-resolution facial photos and identity information of all company employees, including name, employee number, department, position, skill level, and disciplinary records. The system uses artificial intelligence algorithms such as cluster analysis and similarity matching to quickly find the most similar historical facial photos to the captured image and extract the associated identity information. If the matching degree is lower than a preset safety threshold, the system determines that the worker's identity is unknown and rejects their request; if the matching degree is higher than the safety threshold, the system successfully obtains the worker's identity information, records it in the tool collection log database, and proceeds to the next step. It is worth mentioning that if the system detects frequent requests for unauthorized use, it will promptly send an alert to the management team, indicating that there may be unauthorized use.

[0110] After successfully obtaining the worker's identity, the system immediately matches them with the worker's current production task to verify the compliance of their requisition behavior. Firstly, the system compares the worker's skill level with the tooling requirements of the task. If the skill level is lower than the tooling requirements, the system warns the worker that they are not qualified to requisition the corresponding tool and guides them to choose a tool that matches their skill level. Secondly, the system compares the worker's position with the tooling allocation level determined by the task priority. If the allocation level is insufficient, the system informs the worker that they do not currently meet the conditions for requisitioning high-end tools and advises them to requisition ordinary tools according to the alternative plan. Simultaneously, the system also accesses the employee disciplinary record database. If it finds that the worker has a history of improper tool use or unauthorized lending to others, further restrictive measures such as reducing requisition privileges and shortening the requisition period will be taken. Through strict identity authentication and behavior control, the system minimizes human-caused risks in tool use.

[0111] Based on the above embodiments, as an optional embodiment, the allocation level determination module further includes: a job matching module;

[0112] The job matching module is used to match the worker's identity information with data in a preset job information database to determine the process flow of the product the worker is responsible for.

[0113] Specifically, the worker information acquisition module is equipped with high-definition cameras, facial recognition software, and other hardware and algorithm models to monitor the tool collection process in real time. When a worker arrives at the tool collection point and requests tools, the system immediately captures a facial image of the worker using the camera, extracting key feature points such as the coordinates and proportions of the eyes, nose, and mouth to create a unique facial feature vector for that worker. The system then compares this feature vector with a pre-set facial image database. This database stores high-resolution facial photos and identity information of all company employees, including name, employee number, department, position, skill level, and disciplinary records. The system uses artificial intelligence algorithms such as cluster analysis and similarity matching to quickly find the most similar historical facial photos to the captured image and extract the associated identity information. If the matching degree is lower than a preset safety threshold, the system determines that the worker's identity is unknown and rejects their request; if the matching degree is higher than the safety threshold, the system successfully obtains the worker's identity information, records it in the tool collection log database, and proceeds to the next step. It is worth mentioning that if the system detects frequent requests for unauthorized use, it will promptly send an alert to the management team, indicating that there may be unauthorized use.

[0114] After successfully obtaining the worker's identity, the system immediately matches them with the worker's current production task to verify the compliance of their requisition behavior. Firstly, the system compares the worker's skill level with the tooling requirements of the task. If the skill level is lower than the tooling requirements, the system warns the worker that they are not qualified to requisition the corresponding tool and guides them to choose a tool that matches their skill level. Secondly, the system compares the worker's position with the tooling allocation level determined by the task priority. If the allocation level is insufficient, the system informs the worker that they do not currently meet the conditions for requisitioning high-end tools and advises them to requisition ordinary tools according to the alternative plan. Simultaneously, the system also accesses the employee disciplinary record database. If it finds that the worker has a history of improper tool use or unauthorized lending to others, further restrictive measures such as reducing requisition privileges and shortening the requisition period will be taken. Through strict identity authentication and behavior control, the system minimizes human-caused risks in tool use.

[0115] On the other hand, this application also provides a tool management method based on an intelligent tool cabinet, such as... Figure 3 The method includes:

[0116] S101: Obtain tool inventory data and product order data, and determine the product production plan based on the product order data;

[0117] S102. Based on the product production plan, determine the tooling plan, and based on the tooling plan and inventory data, determine the tooling replenishment plan.

[0118] The above are merely exemplary embodiments 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. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the disclosure herein.

[0119] 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. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A tool management system based on an intelligent tool cabinet, characterized in that, The system is applied to an intelligent tool cabinet, and the system includes: a data acquisition module, a data processing module, and a tool replenishment plan generation module; The data acquisition module is used to acquire tool inventory data and product order data; The data processing module is used to determine the product production plan based on the product order data; The tool replenishment plan generation module is used to determine the tool usage plan based on the product production plan, and to determine the tool replenishment plan based on the tool usage plan and the inventory data. The tool replenishment plan generation module includes: a tool supply and demand difference determination module and a tool replenishment plan module; The tool supply and demand difference determination module is used to compare the tool demand data at each time point in the tool usage plan with the inventory data and calculate the tool supply and demand difference at each time point. The tool replenishment planning module is used to form a tiered replenishment plan table with time axis as the baseline and product category as the unit based on the tool supply and demand difference at each time point, the preset supplier delivery cycle and the product production plan, and to determine the tool replenishment plan according to the tiered replenishment plan table. The system also includes: an image acquisition module, a wear degree recognition module, and a tool return and storage module; The image acquisition module is used to acquire images of the returned tools; The wear degree recognition module is used to determine the wear degree of the returned tool based on the image of the tool; The returned tool storage module is used to classify the returned tools according to the degree of wear, obtain the category data of the returned tools, and generate inventory tool category data based on the category data of the returned tools and the inventory data of the tools. The system also includes: an allocation level determination module and a tool allocation module; The allocation level determination module is used to obtain the worker's identity information when the worker receives the cutting tools, determine the process flow of the product the worker is responsible for based on the worker's identity information, and determine the allocation level of the cutting tools based on the process flow of the product the worker is responsible for and the product production plan. The tool allocation module allocates tools according to the tool allocation level and the inventory tool category data.

2. The system according to claim 1, characterized in that, The tool replenishment plan generation module includes a production plan decomposition module, a production process analysis module, a tool consumption prediction module, and a tool usage plan generation module, including: The production plan decomposition module is used to decompose the production plan to obtain production batches, production quantities, and production process flows. The production process analysis module is used to determine the type and quantity of cutting tools required for each process based on the production process flow. The tool consumption prediction module is used to acquire historical tool consumption data and determine the tool consumption amount for each process based on the historical tool consumption data and the type and quantity of tools required for each process. The tool usage plan generation module is used to determine the tool usage plan based on the tool consumption of each process, the production batch, and the production quantity.

3. The system according to claim 1, characterized in that, The wear degree identification module includes: a feature extraction module and a wear identification module; The feature extraction module is used to extract key features through image segmentation algorithms; The wear identification module is used to obtain the wear degree of the returned tool by inputting a preset wear degree evaluation model based on the key features.

4. The system according to claim 1, characterized in that, The allocation level determination module includes: a process flow analysis module, a priority determination module, a tool requirement level determination module, and a tool allocation level determination module; The process flow analysis module is used to perform a detailed analysis of the process flow of the product under the worker's responsibility, and to obtain the processing characteristics, quality requirements, process parameters and key processes of each process. The priority determination module is used to determine the priority of the process flow of the product that the worker is responsible for, based on the product production plan. The tool requirement level determination module is used to determine the tool requirement level based on the processing characteristics of each process, the quality requirements, the process parameters, and the key processes. The tool allocation level determination module is used to determine the allocation level of the tool based on the priority and the tool requirement level.

5. The system according to claim 1, characterized in that, The allocation level determination module further includes: a worker information acquisition module; The worker information acquisition module is used to acquire the worker's facial image when the worker receives the knife, and compare the facial image with a preset image database to obtain the worker's identity information.

6. The system according to claim 1, characterized in that, The allocation level determination module further includes: a job matching module; The job matching module is used to match the worker's identity information with data in a preset job information database to determine the process flow of the product the worker is responsible for.

7. A tool management method based on an intelligent tool cabinet, characterized in that, The method includes: Obtain tool inventory data and product order data, and determine the product production plan based on the product order data; Based on the product production plan, determine the tooling plan, and based on the tooling plan and the inventory data, determine the tooling replenishment plan; Compare the tool demand data at each time point in the tool usage plan with the inventory data, and calculate the tool supply and demand difference at each time point; Based on the tool supply and demand difference at each time point, the preset supplier delivery cycle, and the product production plan, a tiered replenishment plan table is formed with the time axis as the baseline and the product category as the unit. The tool replenishment plan is determined according to the tiered replenishment plan table. Get a picture of the returned knife; The degree of wear on the returned knife is determined based on the image of the knife. The returned tools are classified according to their wear level to obtain category data of the returned tools, and inventory tool category data is generated based on the category data of the returned tools and the inventory data of the tools. When a worker receives a cutting tool, the worker's identity information is obtained, and the process flow of the product the worker is responsible for is determined based on the worker's identity information. Based on the process flow of the product the worker is responsible for and the product production plan, the allocation level of the cutting tool is determined. Tools are allocated based on the tool allocation level and the inventory tool category data.

Images