Needle ring using and needle supplementing management system based on artificial intelligence multi-permission grading

Abstract

This invention discloses a needle requisition and replacement management system based on artificial intelligence and multi-level access control, belonging to the field of intelligent consumable management technology. It includes a permission management module, a requisition management module, a replacement needle prediction module, an inventory management module, and a data interaction module. The permission management module calculates and classifies comprehensive permission values ​​based on job-specific weights, historical requisition compliance, and production task relevance using a permission quantification model. The replacement needle prediction module integrates historical consumption data, production plans, equipment status, and inventory to derive replacement needle demand using a consumption prediction model. The requisition management module calculates the single requisition limit using a dynamic limit adjustment model. The inventory management module executes replacement needles and updates inventory. The data interaction module enables real-time data synchronization between modules and external systems. This system solves problems such as rigid permissions, delayed replacement needles, and fixed limits in traditional management systems, and is suitable for the needs of manufacturing enterprises.

Description

Technical Field

[0001] This invention relates to the field of intelligent management technology for industrial consumables, and specifically to a needle requisition and replacement management system based on artificial intelligence and multi-level access control. Background Technology

[0002] Sewing machine needles are core consumables in the garment and home textile manufacturing industries. Currently, most companies face three major pain points in needle management:

[0003] Rigid access control: Static roles (such as administrators and operators) are commonly used to assign permissions, which are not dynamically linked to users' actual job contributions, historical compliance, and current production workload. This results in a mismatch between permissions and responsibilities, and the coexistence of unauthorized access and inefficient approval processes.

[0004] Lagging replenishment decisions: Replenishment decisions often rely on the personal experience of warehouse staff or simple safety stock thresholds, failing to systematically integrate multi-source data such as historical actual consumption rates, future production plan consumption, and equipment health status, resulting in the dilemma of "production stoppage due to needle breakage" or "inventory backlog".

[0005] Fixed usage limits: User usage limits are mostly fixed values ​​and cannot be dynamically adjusted according to real-time production tasks, individual usage efficiency, and global inventory pressure. This may restrict production and easily lead to waste.

[0006] Some existing smart warehousing solutions attempt to introduce complex AI models such as neural networks for prediction, but they suffer from drawbacks such as uninterpretable models, large training data requirements, and high deployment costs, making them difficult to implement in small and medium-sized manufacturing enterprises that prioritize reliability and cost.

[0007] Therefore, the present invention aims to provide a needle requisition and replacement management system based on artificial intelligence and multi-level permissions, and fundamentally solve the above problems by constructing a closed-loop control mechanism of "permission-prediction-limit". Summary of the Invention

[0008] The purpose of this invention is to provide a needle requisition and replacement management system based on artificial intelligence and multi-level access control, so as to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a needle requisition and replacement management system based on artificial intelligence and multi-level access control, specifically comprising:

[0010] The system includes a permission management module, a requisition management module, a needle replenishment prediction module, an inventory management module, and a data interaction module for synchronizing data between these modules.

[0011] The permission management module is used to calculate the user's comprehensive permission value based on the user's basic job weight, historical compliance of requisition, and relevance to production tasks through a permission quantification model.

[0012] The needle replenishment prediction module is used to calculate the future cycle needle replenishment demand based on historical consumption data, production plans, equipment status, and current inventory through a consumption prediction model.

[0013] The requisition management module is used to calculate the user's single requisition limit based on the current production task, historical requisition efficiency and inventory status, and by receiving the comprehensive permission value and the needle replenishment demand, through a dynamic limit adjustment model.

[0014] The inventory management module is used to perform needle replacement operations and update inventory data;

[0015] The output parameters of the permission quantification model, the consumption prediction model, and the dynamic limit adjustment model are transferred and linked in real time between the permission management module, the needle replenishment prediction module, and the requisition management module through the data interaction module.

[0016] The technical effects and advantages of this invention are as follows:

[0017] 1. This invention abandons the existing technology of dividing permissions based solely on static roles. Instead, it constructs a multi-dimensional permission quantification model through a permission management module. With the basic weight of the job, historical compliance of requisition, and relevance to production tasks as core inputs, it dynamically generates a comprehensive permission value P through dimension correction, product fusion, and normalization. The model is divided into 5 graded permission levels. Historical compliance of requisition not only serves as a correction factor to optimize the relevance between job weight and production tasks, but also sets a warning threshold to trigger permission warnings and downgrade approval priorities. This ensures that permission allocation is both aligned with job responsibilities and matches users' actual requisition behavior and production contributions, effectively avoiding unauthorized requisition and inefficient approval processes, and achieving precise matching of permissions and responsibilities.

[0018] 2. This invention constructs a multi-source data fusion consumption prediction model through a needle replacement prediction module. It systematically integrates historical consumption data from the past n days, planned production volume for the next t period, equipment wear coefficients, and current inventory. The model undergoes three steps: historical consumption rate correction, planned consumption rate calculation, and needle replacement demand verification. It accurately outputs the needle replacement demand Q. When Q > 0, the system automatically generates a needle replacement request and pushes it for approval; when Q ≤ 0, it records inventory redundancy to adjust parameters, achieving four-dimensional collaborative decision-making across "history, future, equipment, and inventory." This completely avoids the dilemma of "needle breakage leading to production stoppage" and "inventory backlog," ensuring production continuity and reducing inventory costs.

[0019] 3. This invention overcomes the limitations of fixed limits in existing technologies by using a dynamic limit adjustment model in the requisition management module. It achieves intelligent adaptation of requisition limits in three steps: First, it calculates the basic limit based on the current production task volume, the unit task requisition coefficient, and the weighted average of the efficiency of the last n requisitions. Second, it corrects the permission level by combining the comprehensive permission value P, so that "the higher the permission, the stronger the limit adaptation." Third, it introduces the inventory adequacy coefficient and the replenishment pressure coefficient as reverse constraints to ensure that the limit is accurately matched with the inventory status and replenishment needs. At the same time, the system presets the maximum requisition limit for each permission level. The final limit is the smaller value between the model calculation value and the maximum limit for each level. This satisfies the differentiated requisition needs of different production tasks and positions, avoids resource waste and production constraints, and achieves real-time dynamic adaptation of requisition limits to actual scenarios. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the module connection of the present invention;

[0021] Figure 2 This is a flowchart of the calculation process for the permission quantification model of this invention;

[0022] Figure 3 This is a flowchart of the calculation process for the consumption prediction model of this invention;

[0023] Figure 4 This is a flowchart of the calculation process for the dynamic quota adjustment model of this invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] This invention provides, for example Figure 1 The needle requisition and replacement management system based on artificial intelligence and multi-level permissions shown includes: permission management module, requisition management module, replacement prediction module, inventory management module, and data interaction module as the data flow hub.

[0026] The data interaction module uses a RESTful API interface to achieve real-time data synchronization and command transmission between various functional modules and between the system and external enterprise information systems (such as production management system, personnel management system, equipment management system, and inventory management system). The preferred communication protocol is HTTPS, and the data transmission uses the AES-256 encryption algorithm to ensure the security and efficiency of data exchange.

[0027] The permission management module collects users' basic job weights, historical requisition compliance, and production task relevance from the enterprise personnel management system, requisition record database, and production management system, respectively. After being processed by the permission quantification model, it generates a DS4 dataset containing a comprehensive permission value P and the corresponding level, and then pushes it to the requisition management module in real time to provide the core basis for adjusting the requisition limit.

[0028] The data sources for the permission management module are: basic job information dataset (DS1, from the enterprise personnel management system), historical requisition record dataset (DS2, from the system requisition record database), and production task association dataset (DS3, from the enterprise production management system). By calculating the comprehensive permission value P, a comprehensive permission value dataset (DS4) containing P and the corresponding permission level is generated. DS4 is pushed to the requisition management module in real time through the data interaction module as the core parameter for adjusting the requisition limit, so as to realize the collaborative control of permissions and limits.

[0029] like Figure 2 The operation process of the permission quantification model shown is as follows:

[0030] Step 1: Dimensional Interval Correction Calculation: Among them, W j As the basic weight of the position, T g For the correlation of production tasks, C y For historical compliance, α is a preset correction coefficient with a value range of [0.1, 0.3] and a default value of 0.2;

[0031] This step is used to eliminate the dimensional differences in job weighting and production relevance, through historical requisition compliance (C y This positively corrects both, strengthens the correlation between data, and makes the quantification of permissions more closely reflect the actual behavior of users; the correction coefficient is determined based on the historical management data of 100 similar companies in the industry, which is obtained through linear regression analysis of the industry management sample;

[0032] Step 2: Product fusion and normalization operations:

[0033] , in, W j C y T g The values ​​of are all in the range of [0,1].

[0034] This step involves multiplying and fusing the data after three-dimensional correction, and normalizing it to the [0,1] interval through a cubic root operation to adapt to the permission level division range and ensure the comparability and rationality of permission values.

[0035] Detailed descriptions and processing logic for each dataset:

[0036] Job Basic Information Dataset (DS1): Originating from the enterprise personnel management system, it contains information such as user job titles and responsibilities. After data extraction and standardization, it generates job basic weights (W). j ), is a constant with a value range of [0.2, 1.0]. The specific value is determined based on the company's organizational structure documents and job descriptions: Administrator position (responsible for system-wide permission configuration and exception handling) W j =1.0, Workshop Supervisor Position (responsible for approving requisitions and allocating resources within the workshop) W j =0.8, Team Leader Position (responsible for reviewing and approving work requests and assigning tasks within the team) W j =0.6, Operator position (responsible only for submitting their own requisition applications) W j =0.4, Warehouse Administrator Position (responsible for inventory counting and needle replacement) W j =0.2; W j 'To adjust the job weights, historical requisition compliance C is used as the basis.' y Positive corrections reflect the principle that "the higher the compliance level, the stronger the adaptability of job permissions," ensuring that permissions match user behavior.

[0037] Historical Requisition Record Dataset (DS2): Originating from the system's requisition record database, this dataset contains information such as the number of requisitions, approval processes, requisition quantities, and consumption registrations for the past three calendar months. After statistical filtering and processing, a historical requisition compliance score (C) is generated. y ), is a variable with a value range of [0,1], calculated as "the number of compliant requisitions in the past 3 calendar months ÷ the total number of requisitions in the same period"; where "compliant requisition" is defined as: a requisition behavior in which the requisition application is approved by the corresponding authority level, the requisition quantity does not exceed the current limit, and the consumption registration is completed as required after requisition; the statistical cycle is automatically triggered, and the data of the previous cycle is updated on the 1st of each month; when C y When the value is less than 0.6, the system automatically triggers a permission warning, pushes the warning information to the corresponding level of administrator, and simultaneously restricts the approval priority of the user's application. y As a correction factor, W is adjusted in conjunction with other factors. j and T g This enables interaction between dimensions.

[0038] Production Task Association Dataset (DS3): Originating from the enterprise production management system, it contains information such as the planned output of the production task currently managed by the user, the total production task of the workshop to which the user belongs, and the consumption quota of the corresponding process. After association calculation processing, the production task association degree (T) is generated. g), is a variable with a value range of [0,1], and is calculated as "the total number of needles required for the current production task of the user ÷ the total number of needles required for the total production task of the workshop in the same period"; where "the total number of needles required for the current production task" is obtained through the production management system interface, and the calculation logic is "the planned output of the current task × the needle consumption quota per unit product of the corresponding process (i.e., k2 in the needle replenishment prediction module)"; T g 'This is the corrected correlation, also determined by C' y The revision reflects the principle that "the better the compliance record of requisition, the higher the weight associated with production tasks," ensuring that permission allocation matches production contributions.

[0039] The comprehensive permission value dataset (DS4) is generated by the permission quantification model after two-step processing of DS1, DS2, and DS3. It includes the comprehensive permission value P and the corresponding permission level for each user. Based on the comprehensive permission value P, five permission levels are defined. The level thresholds are determined through industry management scenario research and calibration based on actual enterprise needs: P∈[0.8,1.0] is Level 1 (highest permission, able to approve all workshop requisition applications, adjust system permission parameters, and view all system requisition and inventory data); P∈[0.6,0.8) is Level 2 (workshop-level permission, able to approve all requisition applications in this workshop, view this workshop's requisition and inventory data, and submit this workshop's replenishment requests); P∈[0.4,0.6) is Level 3 (team-level permission). Permissions include: Level 4 (operational level permission, can submit individual requisition applications and view individual requisition records); Level 5 (restricted permission, requisition applications must be manually reviewed by users with Level 2 or higher permissions, and the number of requisitions per month cannot exceed 3); and DS4 pushes data to the requisition management module in real time through the data interaction module as the core parameter for quota adjustment, realizing data linkage between modules.

[0040] The needle replacement prediction module collects historical consumption data, production plans, equipment status and current inventory from equipment sensors, production management system, equipment management system and inventory system respectively, and calculates the needle replacement demand Q after analysis by the consumption prediction model.

[0041] The required needle replacement quantity is used to generate a DS9 dataset. This dataset is pushed to the inventory management module to trigger the needle replacement application process, and to the requisition management module to provide data support for the dynamic constraint of requisition limits.

[0042] The data sources for the needle replenishment prediction module include: historical consumption statistics dataset (DS5, sourced from the equipment IoT sensor acquisition system and requisition record database), production plan information dataset (DS6, sourced from the enterprise production management system), equipment status monitoring dataset (DS7, sourced from the enterprise equipment management system), and real-time inventory dataset (DS8, sourced from the enterprise inventory management system). By calculating the needle replenishment demand quantity Q, a needle replenishment demand dataset (DS9) is generated. DS9 is synchronously pushed to the inventory management module and the requisition management module through the data interaction module to realize the linkage between needle replenishment triggering and requisition limit constraints.

[0043] like Figure 3 The consumption prediction model shown below has the following specific steps:

[0044] Step 1: Basic Consumption Rate Fusion Calculation:

[0045] , where β is the preset loss correction coefficient, with a value range of [0.05, 0.2], and a default value of 0.1;

[0046] This step integrates historical consumption data from the past n days and introduces an equipment loss coefficient (k3) to correct the consumption rate, making the historical consumption data more consistent with the actual operating conditions of the equipment; the correction coefficient β is obtained based on the variance analysis of consumption data of equipment with different service years.

[0047] Step Two: Production Plan Consumption Calculation:

[0048] Q s Let k1 be the planned production quantity for the next period t, k2 be the unit product consumption quota, k3 be the equipment loss coefficient, and R be the production quantity for the next period t. s The revised planned consumption rate R s ;

[0049] This step combines future production plans with process consumption quotas and simultaneously uses equipment loss coefficient correction to ensure that planned consumption forecasts match the actual operating status of equipment and improve the forecast's foresight.

[0050] Step 3: Calculation of Needle Replacement Requirement:

[0051] Q=(R×k1+R s ×(1-k1))×mS, where m is the number of currently running devices, S is the current inventory, and k1 is the fusion weight, with a value range of [0.4, 0.8] and a default value of 0.6;

[0052] This step integrates historical and planned corrective consumption, and combines the current inventory quantity to determine the replenishment demand, achieving a three-dimensional balance of "history-future-inventory" to avoid untimely replenishment or inventory backlog.

[0053] Detailed descriptions and processing logic for each dataset:

[0054] Historical Consumption Statistics Data Set (DS5): Sourced from the device's IoT sensor acquisition system and usage record database, it includes information such as needle replacement records and manually supplemented consumption data from the past n days. After aggregation and statistical processing, it generates the average historical consumption rate for the past n days. n represents the number of days for statistics, with a default value of 30 days, which can be adjusted by the administrator according to the production cycle (adjustment range: 15-90 days). The actual number of needles consumed on day i-1 is collected in real time, and the daily consumption data is summarized at 24:00 every day. After the first step of the model calculation, the corrected historical consumption rate (R) is generated, with the unit being "needles / day", which reflects the impact of equipment wear and tear on historical consumption.

[0055] Production Planning Information Dataset (DS6): Originating from the enterprise production management system, it contains planned production output for the next t-period, corresponding process information, etc. After process matching and quantification, it generates the planned production quantity for the next t-period (Q). s The model calculates the production plan consumption rate (R) and the unit product needle consumption quota (k2). t is the forecast period in days, entered by the user in the system or automatically obtained through the production management system API interface, with an entry / retrieval frequency of once per day. k2 is a constant, set according to the differences in production processes, and its value is calibrated based on the industry standard "Textile Machinery Needle Consumption Quota" (FZ / T90115-2023) and the company's actual production data. Example values: flat sewing process k2 = 0.02 needles / piece, overlock sewing process k2 = 0.03 needles / piece, buttonhole process k2 = 0.05 needles / piece. After the second step of the model calculation, the corrected production plan consumption rate (R) is generated. s The unit is "roots / day", which reflects the impact of equipment wear and tear on planned consumption.

[0056] Equipment Status Monitoring Dataset (DS7): Sourced from the enterprise equipment management system, it includes information such as equipment model, acceptance and commissioning time, and current operating status. After status filtering and age-based statistical processing, it generates the number of currently operating equipment (m) and the equipment wear coefficient (k3). m is a variable, with the unit being "units," and is acquired once per hour. Only equipment in "running" or "standby" status is counted, excluding equipment in "maintenance" or "discontinued" status. k3 is a constant, divided according to the actual service life of the equipment, with values ​​determined based on the equipment manual and industry experience data: equipment service life ≤ 1 year (good equipment condition, minimal wear) k3 = 1.0, 1-3 years (normal equipment condition, moderate wear) k3 = 1.2, ≥ 3 years (deteriorating equipment condition, significant wear) k3 = 1.5. The equipment service life is calculated from the date the equipment is accepted and put into use, and is automatically updated by the equipment management system. DS7 is used as a correction parameter in the first two steps of the model calculation to adapt the consumption rate to the equipment status.

[0057] The real-time inventory dataset (DS8) is derived from the enterprise inventory management system and includes information such as the current inventory quantity of needles, inbound and outbound records, and inventory count data. After real-time collection and calibration processing, the current inventory quantity of needles (S) is generated. S is a variable, and the unit is "needle". Inbound and outbound records are synchronized in real time. A physical inventory count is performed at the end of each month, and the inventory count data is used to calibrate the system inventory data. After DS8, DS5, DS6, and DS7 are processed by model calculations, the replacement needle demand dataset (DS9) is generated.

[0058] The needle replenishment demand dataset (DS9) is generated by the consumption prediction model after processing DS5, DS6, DS7, and DS8. It contains the needle replenishment demand Q for the future period t, with the unit being "needles". DS9 is pushed to the inventory management module and the requisition management module in real time through the data interaction module. When Q>0, the inventory management module automatically generates a needle replenishment application form and pushes it to users with Level 2 or higher permissions for approval. At the same time, the inventory adequacy coefficient k5 is lowered, which inversely constrains the requisition limit and realizes linkage with the requisition management module. When Q≤0, no needle replenishment is required, and the system records the current inventory redundancy for reference in subsequent limit adjustments.

[0059] The requisition management module is used to calculate the user's single requisition limit based on the current production task, historical requisition efficiency and inventory status, and by receiving the comprehensive permission value and the needle replenishment demand, through a dynamic limit adjustment model.

[0060] The data sources for the requisition management module include: the current production task dataset (DS10, from the enterprise production management system), the historical requisition efficiency dataset (DS11, from the system requisition record database), the inventory adequacy dataset (DS12, from the enterprise inventory management system), as well as DS4 output by the permission management module and DS9 output by the needle replenishment prediction module; the single requisition limit L is calculated to generate the single requisition limit dataset (DS13); a dual verification mechanism of "permission level + dynamic limit" is adopted. When a user submits a requisition application, the matching of the permission level and the approval process is verified first, and then the consistency of the requisition quantity with DS13 is verified. If any condition is not met, the approval is rejected or upgraded.

[0061] like Figure 4 The calculation steps for the dynamic quota adjustment model shown are as follows:

[0062] Step 1: Calculation of Basic Limit:

[0063] Based on the current production task Q t Unit task allocation coefficient k4 and near The weighted average of the requisition efficiency is used to calculate the basic limit L0. , Let i be the quantity used in the (i-1)th time. The efficiency of the (i-1)th requisition;

[0064] This step involves determining the initial requisition limit by combining the current production workload with historical requisition efficiency, ensuring that the limit matches production needs and user requisition habits.

[0065] Step 2: Permission Level Correction Calculation

[0066] Based on the basic limit L0 and the comprehensive permission value P, the permission level is adjusted to obtain the first adjusted limit L1. Where γ and δ are preset permission correction coefficients, with γ ranging from [0.7, 0.9] and a default value of 0.8, and δ ranging from [0.3, 0.5] and a default value of 0.4;

[0067] This step introduces the comprehensive permission value (P) of the permission management module to positively correct the limit, realizing "the higher the permission, the stronger the limit adaptability", matching the enterprise's multi-level management needs; the correction formula coefficient is calibrated through enterprise permission and limit matching samples. The coefficient after correction for Level 4 permission (P≈0.3) is 0.92, and the coefficient after correction for Level 2 permission (P≈0.7) is 1.08.

[0068] Step 3: Inventory and Replenishment Demand Adjustment Calculation:

[0069] Based on the first correction limit L1, the inventory adequacy coefficient k5, and the replenishment pressure coefficient, the final limit L is obtained by adjusting the inventory and replenishment demand. Where ε and λ are the preset inventory pressure correction coefficients and upper limits, respectively. The value of ε ranges from [0.1, 0.3], with a default value of 0.2, and the value of λ is 0.5. This is the floor function;

[0070] This step introduces an inventory adequacy coefficient (k5) and a reverse constraint limit on the replenishment demand (Q) of the replenishment forecast module to avoid excessive requisition when inventory is tight or replenishment pressure is high; the inventory pressure correction coefficient and upper limit are determined based on industry inventory management experience data.

[0071] Detailed descriptions and processing logic for each dataset:

[0072] Current Production Task Dataset (DS10): Originating from the enterprise production management system, it contains the total number of production tasks currently assigned by the user, corresponding process information, etc. After task matching processing, it generates the current production task quantity (Q). t ) and unit task allocation coefficient (k4); Q t k is a variable, with the unit being "pieces", and the acquisition frequency is once every half day; k4 is a constant, corresponding one-to-one with k2 in the replacement stitch prediction module, and its value is 1.1 times that of k2 (with 10% loss redundancy reserved). Example values: k4 = 0.022 stitches / piece for flat sewing, k4 = 0.033 stitches / piece for overlock sewing, and k4 = 0.055 stitches / piece for buttonhole sewing.

[0073] Historical requisition efficiency dataset (DS11): derived from the system requisition record database, containing recent... Information such as the number of users who received and the actual amount consumed is used to generate a near-perfect database after efficiency calculations and anomaly removal. Weighted average of requisition efficiency ; Let i be the quantity used in the (i-1)th time. The efficiency of the (i-1)th requisition is calculated as "the actual quantity consumed after the (i-1)th requisition ÷ the quantity requisitioned in the (i-1)th requisition", with a value range of (0, 1.2]. When the value is >1.2), it is judged as an abnormal requisition, and the data is not included in the statistics. After the first step of the model calculation, DS10 and DS11 generate the uncorrected basic limit (L0), with the unit being "root", which reflects the basic impact of production tasks and historical requisition efficiency on the limit.

[0074] The inventory adequacy dataset (DS12) originates from the enterprise inventory management system and includes information such as the current inventory quantity S and the safety stock quantity S0. After adequacy calculation, an inventory adequacy coefficient (k5) is generated. S0 is calculated as "the average daily maximum consumption rate over the past 3 months × 7 days (safety reserve period)". S0 is calibrated monthly. The specific values ​​of k5 are: S / S0≥2 (sufficient inventory, limit can be appropriately increased) k5=1.2, 1≤S / S0<2 (normal inventory, maintain the baseline limit) k5=1.0, S / S0<1 (tight inventory, limit should be appropriately reduced) k5=0.8. DS12, combined with DS4 (including P value) output by the permission management module and DS9 (including Q value) output by the needle replenishment prediction module, is processed by the model to generate the corrected limit and the final limit.

[0075] The single-use limit dataset (DS13) is generated by the dynamic limit adjustment model after processing DS10, DS11, DS12, DS4, and DS9 through three steps. It includes the user's single-use limit L and the maximum limit for the corresponding permission level. L is converted to an integer by rounding down to ensure that it conforms to the actual use scenario. The maximum limit for the permission level is a constant, set according to the enterprise's resource management needs and the industry average. The specific values ​​are: Level 1 (administrator) maximum limit is unlimited, Level 2 (workshop supervisor) maximum limit ≤ 500 units / use, Level 3 (team leader) maximum limit ≤ 200 units / use, Level 4 (operator) maximum limit ≤ 50 units / use, and Level 5 (restricted permission) maximum limit ≤ 20 units / use. The final use limit L must simultaneously meet the model calculation result and the maximum limit for the corresponding permission level, taking the smaller value of the two to ensure that the permission control does not fail.

[0076] The present invention will be further described in detail below with reference to specific embodiments:

[0077] Example 1: Application of a needle management system in a textile factory

[0078] The textile factory has three production workshops, each equipped with 20 sewing machines (all JUKIDDL-8700 models, with the following service lifespans: Workshop 1: 1.5 years, Workshop 2: 2.5 years, Workshop 3: 0.8 years). The processes include flat sewing and overlock sewing. The factory currently has one administrator, three workshop supervisors, six team leaders, 50 operators, and two warehouse managers. The system deployment adopts a B / S architecture, with the backend developed in Java, the database using MySQL 8.0, and the frontend using the Vue 3.0 framework. Each module interacts via a RESTful API interface, using HTTPS 1.2 as the communication protocol and AES-256 as the data transmission encryption algorithm. The interface is called once per hour to ensure real-time data transmission. Those skilled in the art can complete the system setup based on the above deployment parameters without additional creative effort.

[0079] Typical Scenario Example (Example 1):

[0080] Authorization hierarchy (authorization quantification model): Taking the workshop 2 manager (position: workshop manager, W) as an example. j Taking 0.8 as an example, the system obtains DS1 from the personnel management system and determines W. j =0.8; Retrieve DS2 from the requisition record database and count the requisition records for the past 3 calendar months: a total of 32 requisitions, of which 30 were compliant requisitions, and calculate C. y =30 / 32≈0.94; Obtain DS3 from the production management system. The current planned production output is 2000 pieces, corresponding to the overlock sewing process (k2=0.03 needles / piece). The current needle count for the task is 2000×0.03=60 needles. The total needle count for the concurrent production task in workshop 2 is 200 needles. Calculate T. g =60 / 200=0.3;

[0081] Dimensional correction: W j '=0.8×(1+0.2×0.94)=0.8×1.188≈0.9504, T g =0.3×(1+0.2×0.94)=0.3×1.188≈0.3564;

[0082] Normalization: P = = ≈0.68, generate the corresponding permission information for the user in DS4, corresponding to Level 2 permissions; the system pushes DS4 to the requisition management module in real time through the data interaction module, and at the same time updates the user's permission level identifier in the permission management interface.

[0083] Needle replenishment forecast (consumption forecast model): Predicts the demand for needle replenishment over the next 7 days (t=7). The system retrieves DS8 from the inventory management system, with current inventory S=800 needles; DS7 from the equipment management system, with current operating equipment m=60 units (all equipment in 3 workshops is in operation), k3=1.2 for workshop 1 equipment, k3=1.2 for workshop 2 equipment, and k3=1.0 for workshop 3 equipment, with a weighted average k3=(20×1.2+20×1.2+20×1.0) / 60≈1.133; DS5 from the consumption record database, with an average historical consumption rate of 20 needles / day over the past 30 days; and DS6 from the production management system, with a planned production quantity Q for the next 7 days. s =5000 pieces (3000 pieces for flat sewing, k2=0.02 threads / piece; 2000 pieces for overlock sewing, k2=0.03 threads / piece), weighted average k2=(3000×0.02+2000×0.03) / 5000=0.12 / 5000=0.024 threads / piece; take k1=0.6;

[0084] Historical consumption correction:

[0085] R = 20 × (1 + 0.1 × (1.133 - 1)) = 20 × 1.0133 ≈ 20.266 roots / day;

[0086] Production plan consumption:

[0087] R s =5000×0.024×(1+0.1×(1.133-1))=120×1.0133≈121.596 roots / day;

[0088] Needle replacement verification:

[0089] Q = (20.266 × 0.6 + 121.596 × 0.4) × 60 - 800 = (12.1596 + 48.6384) × 60 - 800 = 60.798 × 60 - 800 = 3647.88 - 800 = 2847.88 needles. Rounding up to 2848 needles, DS9 is generated (Q = 2848 needles, triggering a needle replacement). The system automatically generates a needle replacement request form, pushes it to the workshop supervisor for approval, and simultaneously pushes DS9 to the requisition management module in real time.

[0090] Issuance limit (limit adjustment model): Workshop 2 operator (Level 4 permission, P=0.3 in DS4) submits an issuance request. The system retrieves DS10 from the production management system, and the current production task quantity Q. t=1000 pieces (overlock sewing process, k4=0.033 stitches / piece); retrieve DS11 from the requisition record database, the weighted average efficiency of the last 10 requisitions = 45 stitches; retrieve DS12 from the inventory management system, current inventory S=800 stitches, safety stock S0=the maximum daily consumption rate of the last 3 months × 7=30×7=210 stitches, S / S0=800 / 210≈3.81, k5=1.2; receive DS4 (P=0.3) pushed by the permission management module and DS9 (Q=2848 stitches) pushed by the replenishment prediction module.

[0091] Basic limit:

[0092] L0 = 1000 × 0.033 + 45 = 33 + 45 = 78 pieces;

[0093] Permission correction:

[0094] L1 = 78 × (0.8 + 0.4 × 0.3) = 78 × 0.92 = 71.76 roots;

[0095] Needle replacement and inventory adjustment:

[0096] The pressure coefficient of the replacement needle = min(2848 / 800, 0.5) = min(3.56, 0.5) = 0.5, L = 71.76 × 1.2 × (1 - 0.2 × 0.5) = 71.76 × 1.2 × 0.9 = 77.4912 =77 roots; Since the maximum limit for Level 4 permissions is 50 roots / time, DS13 is generated, and the final limit is determined to be 50 roots; The system displays on the application interface that the user can apply for a maximum of 50 roots at a time. The user enters the number of roots to be applied for, which is 45. The system verifies and pushes it to the team leader for approval.

[0097] Example of abnormal scenario handling (Example 2):

[0098] Scenario 1: User's historical usage compliance C y Early warning and control of <0.6

[0099] A workshop operator (Level 4 access, original P=0.35) made a total of 25 requisitions in the past 3 calendar months, with 12 of them being compliant requisitions. Calculate C. y =12 / 25=0.48<0.6, triggering a permission warning.

[0100] System Actions: 1. Automatically generate early warning information and push it to the team leader (Level 3) and workshop supervisor (Level 2) of the relevant team. The early warning content includes C. yNumerical values ​​and details of abnormal requisition records;

[0101] 2. The permission management module recalculates its overall permission value P: W j =0.4, T g =0.25, correction:

[0102] W j =0.4×(1+0.2×0.48)=0.4×1.096=0.4384;

[0103] T g =0.25×(1+0.2×0.48)=0.25×1.096=0.274;

[0104] Normalization: P = = ≈0.38, still Level 4, but the priority of requisition and approval has been lowered;

[0105] 3. The requisition management module receives the updated DS4 and adjusts its requisition limit: the original L = 50 units / time, the revised L = ⌊original L1×k5×(1-0.2×min(Q / S,0.5))⌋ = 35 units / time, and the monthly requisition limit is 5 times; 4. The system records the user's warning status and recalculates C on the 1st of each month. y C for three consecutive months y If the value is ≥0.6, the warning will be lifted and the original permissions and limits will be restored.

[0106] Scenario 2: Redundancy handling and forecast adjustment when the replacement needle demand Q is negative

[0107] When predicting the demand for replacement needles in the next 7 days, the system obtains DS8: current inventory S=3500 needles; DS7: m=60 units, k3=1.133; DS5: average historical consumption rate in the past 30 days=20 needles / day; DS6: production plan Qs for the next 7 days=1000 units (k2=0.024 needles / unit); k1=0.6.

[0108] Step 1: R = 20 × (1 + 0.1 × (1.133 - 1)) ≈ 20.266 roots / day;

[0109] Step Two: R s =1000×0.024×(1+0.1×(1.133-1))≈24.319 roots / day;

[0110] Step 3:

[0111] Q = (20.266 × 0.6 + 24.319 × 0.4) × 60 - 3500 = (12.1596 + 9.7276) × 60 - 3500 = 21.8872 × 60 - 3500 = 1313.232 - 3500 = -2186.768 roots, Q < 0.

[0112] System action: Record the inventory redundancy of 2187 units (absolute value), store it in the redundancy record database of the inventory management module, and mark the reason for redundancy as: production plan reduction;

[0113] The needle replacement prediction module automatically adjusts subsequent prediction parameters, reducing the fusion weight k1 from 0.6 to 0.4, thereby increasing the influence weight of the production plan;

[0114] The requisition management module receives DS9 (Q=-2187), the needle replacement pressure coefficient = min(-2187 / 3500, 0.5) = 0, the limit correction coefficient = 1 - 0.2 × 0 = 1, and appropriately increases the requisition limit to encourage reasonable requisition and digestion of redundancy;

[0115] At the end of each month, the system compiles statistics on the redundancy reduction and adjusts the safety stock S0 based on subsequent production plans.

[0116] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A needle requisition and replacement management system based on artificial intelligence and multi-level access control, characterized in that: include: The system includes a permission management module, a requisition management module, a needle replenishment prediction module, an inventory management module, and a data interaction module for synchronizing data between these modules. The permission management module is used to calculate the user's comprehensive permission value based on the user's basic job weight, historical compliance of requisition, and relevance to production tasks through a permission quantification model. The needle replenishment prediction module is used to calculate the future cycle needle replenishment demand based on historical consumption data, production plans, equipment status, and current inventory through a consumption prediction model. The requisition management module is used to calculate the user's single requisition limit based on the current production task, historical requisition efficiency and inventory status, and by receiving the comprehensive permission value and the needle replenishment demand, through a dynamic limit adjustment model. The inventory management module is used to perform needle replacement operations and update inventory data; The output parameters of the permission quantification model, the consumption prediction model, and the dynamic limit adjustment model are transferred and linked in real time between the permission management module, the needle replenishment prediction module, and the requisition management module through the data interaction module.

2. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 1, characterized in that, The permission quantification model calculates the user's comprehensive permission value P through the following steps: Step 1: Based on historical requisition compliance C y Basic weight W for the position j and production task correlation T g Make corrections to obtain the corrected job weight W. j The correlation between ′ and the revised production task T g ′, in α is a preset correction coefficient; Step 2: Based on W j ′、C y and T g Calculate the overall permission value P. , Among them, W j C y T g The value range is [0,1].

3. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 2, characterized in that, The system divides user permissions into multiple levels based on the comprehensive permission value P; when the historical usage compliance C y When the threshold is lower than the preset threshold, the system triggers a permission warning and lowers the priority of the corresponding user's access approval.

4. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 1, characterized in that, The consumption prediction model calculates the needle replacement demand Q in the future period t through the following steps: Step 1: Based on the historical consumption data of the past n days and the equipment loss coefficient k3, calculate the corrected historical consumption rate R. Where β is the preset loss correction coefficient, and n is the number of statistical days. This represents the actual number of needles consumed on day i-1. Step 2: Production plan Q based on future period t s Calculate the corrected planned consumption rate R based on the unit product consumption quota k2 and the equipment loss coefficient k3. s , ; Step 3: Based on R, R s Given the current number of operating devices m, the current inventory S, and the fusion weight k1, calculate the replacement needle demand Q, Q = (R × k1 + R s ×(1-k1))×mS.

5. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 4, characterized in that, When the required quantity of replacement needles, Q, is greater than zero, the inventory management module automatically generates a replacement needle request form; when Q is less than or equal to zero, the system records the inventory redundancy and uses it for subsequent parameter adjustments.

6. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 1, characterized in that, The dynamic limit adjustment model calculates the user's single-use limit L through the following steps: Step 1: Based on the current production task quantity Q t Unit task allocation coefficient k4 and near Weighted average of requisition efficiency , Let i be the quantity used in the (i-1)th time. To determine the efficiency of the (i-1)th requisition, calculate the basic limit L0. ; Step 2: Based on the basic limit L0 and the comprehensive permission value P, perform permission level correction to obtain the first corrected limit L1. , where γ and δ are preset permission correction coefficients; Step 3: Based on the first adjusted limit L1, the inventory adequacy coefficient k5, and the replenishment pressure coefficient, adjust the inventory and replenishment demand to obtain the final limit L. Where ε is the preset inventory pressure correction coefficient and λ is the upper limit value. This is the floor function.

7. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 6, characterized in that, The system presets a maximum usage limit for each permission level; the value of the final limit L must not exceed the maximum usage limit preset for the corresponding user permission level.

8. The needle requisition and replacement management system based on artificial intelligence and multi-level access control as described in claim 1, characterized in that, The data interaction module connects with external production management systems, personnel management systems, equipment management systems, and inventory management systems via RESTful API interfaces, with a data synchronization frequency of no less than once per hour.

Images