Information management system and manufacturing management method for intelligent manufacturing of yoga clothes

Abstract

This invention provides an information management system and manufacturing management method for intelligent manufacturing of yoga clothing, belonging to the field of manufacturing management technology. Specifically, it includes: determining the changes in equipment parameters of different production equipment based on monitoring data of the equipment parameters; determining an adjustment scheme for the production equipment based on the changes in the equipment parameters of different production equipment; adjusting the equipment parameters of the production equipment when changes occur using the adjustment scheme; and determining the need for active adjustment targets in the production equipment based on the changes in the recognition data of the image model under different quality defect types and the number of recognition deviations under different quality defect types. This improves the reliability of the image model in recognizing different types of quality defects.

Description

Technical Field

[0001] This invention belongs to the field of manufacturing management technology, and in particular relates to an information management system and manufacturing management method for intelligent manufacturing of yoga clothing. Background Technology

[0002] Currently, the detection of appearance defects in yoga clothing mainly relies on manual visual inspection, which is inefficient and prone to missed defects due to fatigue. In the invention patent application CN202310417996.5, "An Image Defect Detection Method Based on Data Augmentation and Normalized Flow," a normalized flow model with fusion attention mechanism is combined to achieve defect detection in fabric and clothing images. The improved normalized flow model with fusion attention mechanism achieves accurate probability density estimation of the image's latent space, thus obtaining the defect detection results for fabric and clothing images. This significantly improves detection performance, but it has the following technical drawbacks: In the manufacturing of yoga clothing, existing technologies often keep the equipment parameters of the production equipment within an optimal range. This makes it difficult for the model to comprehensively identify different types of quality defects, thus hindering the determination of the model's true reliability. Therefore, it is crucial to determine an adjustment and control scheme for the equipment parameters of the production equipment when they change, based on the defect identification deviation of the image model under different quality defect types and the variation of equipment parameters. This would allow the production equipment to operate within a wider range of equipment parameters, thereby improving the verification reliability of the image model's defect identification deviation under different quality defect types and providing data support for the iterative updating of the image model. This is a pressing technical problem that needs to be solved.

[0003] Therefore, there is an urgent need for an information management system and manufacturing management method for intelligent manufacturing of yoga clothing. Summary of the Invention

[0004] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a manufacturing management method, which includes: S1 uses the defect identification data from the image model to determine the defect identification deviation under the current production parameters. Based on the defect identification deviation under different quality defect types, it determines when production parameter adjustment and control management is needed, and then proceeds to the next step. S2 determines the changes in the equipment parameters of different production equipment based on the monitoring data of the equipment parameters of the production equipment, and determines the adjustment plan for the production equipment based on the changes in the equipment parameters of different production equipment. S3 utilizes the aforementioned adjustment scheme to adjust the equipment parameters of the production equipment when changes occur. Based on the changes in the identification processing data of the image model under different quality defect types, and the number of identification deviations under different quality defect types, when it is determined that active adjustment targets in the production equipment need to be identified, the active adjustment targets in the production equipment are determined based on the changes in the equipment parameters of the production equipment and the adjustment scheme.

[0005] Furthermore, the defect identification data includes the number of identification processes and the identification process results under different quality defect types.

[0006] Furthermore, the defect identification deviation includes the number of identification deviations under different quality defect types.

[0007] Furthermore, it was determined that production parameters needed to be adjusted and controlled, specifically including: Based on the defect identification deviation under different quality defect types, determine the number of identification deviations under different quality defect types; The identification deviation probability of the quality defect type is determined by using the number of identification deviations. Based on the identification deviation probability and the number of identification processes for different types of quality defects, determine whether it is necessary to adjust and control the production parameters.

[0008] Furthermore, the identification deviation probability of the quality defect type is determined based on the proportion of identification deviations in the number of identification processing steps for the quality defect type.

[0009] Furthermore, the variation of the equipment parameters of the production equipment is determined based on the time period during which the variation rate of the equipment parameters of the production equipment meets the requirements.

[0010] Furthermore, the time period in which the variation rate of the equipment parameters of the production equipment meets the requirements is the time period in which the equipment parameters of the production equipment are not in the equipment parameter range corresponding to the reference equipment parameters, and the absolute value of the minimum value of the deviation rate between the average value of the equipment parameters and the endpoint of the equipment parameter range corresponding to the reference equipment parameters of the production equipment is greater than the preset deviation rate threshold.

[0011] Furthermore, the baseline equipment parameters of the production equipment are determined based on the recommended values ​​of the baseline equipment parameters of the production equipment, that is, the optimal equipment parameter range corresponding to the production equipment determined based on historical quality inspection data, while the qualified range is the equipment parameter range in which the production equipment must operate.

[0012] Furthermore, the method for determining the adjustment plan for the production equipment is as follows: Based on the variation of equipment parameters of different production equipment, determine the time period in which the variation rate of the equipment parameters of the production equipment meets the requirements; The time period during which the rate of change of the equipment parameters of the production equipment meets the requirements is taken as the change matching time period of the production equipment. Based on the different time periods for the changes in production equipment, a change adjustment plan for the production equipment is determined.

[0013] Secondly, this invention provides an information management system for intelligent manufacturing of yoga clothing, employing the aforementioned manufacturing management method, specifically including: Adjustment control module, change adjustment module, adjustment target determination module; The adjustment and control module is responsible for determining whether production parameters need to be adjusted and managed. The change and adjustment module is responsible for determining the change and adjustment plan for the production equipment. The adjustment target determination module is responsible for determining the active adjustment target in the production equipment.

[0014] The beneficial effects of this invention are as follows: Based on the model's defect identification deviation under different quality defect types in yoga clothing, the identification deviation risk of the model for different quality defect types is determined. The identification deviation risk is then used to determine whether production parameter adjustment and control management is required, i.e., whether adjustment and control management is required when the equipment parameters of the production equipment change. This ensures that the model can identify and process different quality defect types more frequently, laying the foundation for further iterative processing of the model.

[0015] Based on the data on changes in the equipment parameters of the production equipment and the production equipment with periods of parameter changes, the changes in the equipment parameters of the production equipment under the current adjustment plan are determined. That is, the more production equipment with parameter changes and the more periods of parameter changes of the production equipment with parameter changes, the higher the reliability of the changes. In other words, the more opportunities the model has to identify and process different types of quality defects, and the higher the verification reliability. Using the verification reliability, it is determined which production equipment should be used as the active adjustment target. That is, the production equipment that actively adjusts its equipment parameters when there are no periods of parameter changes in the most recent preset time period. This also lays the foundation for further improving the comprehensiveness and reliability of the model's verification processing.

[0016] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0019] Figure 1 A flowchart of a manufacturing management method; Figure 2 It is a flowchart for determining the methods for adjusting and controlling production parameters that need to be managed; Figure 3 This is a flowchart illustrating the method for determining the adjustment plan for production equipment. Figure 4 This is a framework diagram of an information management system for the intelligent manufacturing of yoga apparel. Detailed Implementation

[0020] 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 specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0021] Example 1 like Figure 1 As shown, this application provides a manufacturing management method, specifically including: S1 uses the defect identification data from the image model to determine the defect identification deviation under the current production parameters. Based on the defect identification deviation under different quality defect types, it determines when production parameter adjustment and control management is needed, and then proceeds to the next step. The defect identification data refers to the identification processing data generated by the image model during the quality inspection of the product in the yoga wear production process, including the number of identification processing times and the identification processing results under different quality defect types; the quality defect types refer to various quality problems that may occur in yoga wear production, including loose fabric fibers, abnormal color difference, seam deviation, insufficient elasticity, and pattern printing deviation; the defect identification deviation refers to the situation where the identification results of the image model are inconsistent with the actual quality inspection results when identifying different quality defect types, specifically manifested as the number of identification deviations; the production parameters refer to the key parameters that control the operating status of the production equipment, including equipment parameters such as temperature, pressure, speed, and tension.

[0022] Suppose a yoga apparel production line is equipped with an image model for real-time quality inspection of products, covering multiple defect types. By statistically analyzing the number of times each defect type is identified and processed, and the number of identification errors, over a certain period, the model's accuracy in identifying different defect types is analyzed. A preset deviation probability threshold, a preset risk defect type quantity threshold, and a preset deviation weight threshold are set. Based on the identification deviation probability, the number of risk defect types identified, and the deviation weight values, it is determined whether production parameter adjustments and control management are necessary.

[0023] This step establishes a complete technical chain from image model defect identification data to adjustment and control management decisions. Its significance lies in systematically evaluating the risk of model identification deviation under various quality defect types, timely identifying quality defect types with insufficient model identification accuracy, thereby triggering the adjustment and control management of production parameters, ensuring that the model can obtain identification training opportunities under more production parameter conditions, and providing a data foundation for the iterative optimization of the model.

[0024] Furthermore, the defect identification data includes the number of identification processes and the identification process results under different quality defect types.

[0025] Specifically, the defect identification deviation includes the number of identification deviations under different quality defect types.

[0026] It is understandable that, such as Figure 2 As shown, it is determined that production parameters need to be adjusted and controlled, specifically including: In this embodiment, based on the defect identification deviation of the model under different quality defect types of yoga clothing, the identification deviation risk of the model for different quality defect types is determined, and the identification deviation risk is used to determine whether production parameter adjustment and control management is required, that is, whether adjustment and control management is required when the equipment parameters of the production equipment change, so as to ensure that the model can identify and process different quality defect types more frequently, laying the foundation for further iterative processing of the model.

[0027] S11 determines the number of identification deviations under different quality defect types based on the defect identification deviations under different quality defect types; The number of identification deviations refers to the cumulative number of times the image model's identification result is inconsistent with the actual quality inspection result when it identifies a certain type of quality defect. The defect identification deviation situation includes the number of identification deviations under different quality defect types, which is used to reflect the accuracy level of the model in identifying each type of quality defect.

[0028] Assuming the image model covers multiple quality defect types, within a certain statistical period, the number of times the model's recognition result differs from the actual quality inspection result for each quality defect type is counted, forming a dataset of recognition deviation counts for each quality defect type, which serves as the basis for subsequent calculation of recognition deviation probability.

[0029] This step, by statistically analyzing the number of identification deviations, quantifies the identification deviation of the image model into comparable values, providing an accurate data foundation for subsequent calculations of identification deviation probabilities and ensuring the objectivity and repeatability of the assessment of the model's identification deviation risk.

[0030] S12 uses the number of identification deviations to determine the identification deviation probability of the quality defect type; The identification deviation probability is determined based on the proportion of identification deviations in the number of identification processing times for the quality defect type, that is, the identification deviation probability is equal to the number of identification deviations for the quality defect type divided by the number of identification processing times for the quality defect type; the number of identification processing times refers to the total number of times the image model performs identification processing for a certain quality defect type within the statistical period.

[0031] Suppose that for a certain type of quality defect, the image model performs a certain number of recognition processes within a statistical period, during which a certain number of recognition errors occur. The recognition error probability of this type of quality defect is calculated by dividing the number of recognition errors by the total number of recognition processes. The above calculation is repeated for all types of quality defects to form a recognition error probability dataset.

[0032] This step, by calculating the probability of identification deviation, is significant because it transforms the number of identification deviations for different quality defect types into a probabilistic form, eliminating the incomparability caused by the different number of identification and processing times for each quality defect type. This makes the identification deviation risks between different quality defect types comparable, providing standardized evaluation indicators for subsequent adjustment, control, and management decisions.

[0033] S13 determines whether production parameter adjustment and control management is required based on the identification deviation probability and the number of identification processes for different quality defect types.

[0034] The preset deviation probability threshold is a critical value used to determine whether the identification deviation probability of a certain quality defect type is too high; the identified risk defect type refers to a quality defect type whose identification deviation probability is greater than the preset deviation probability threshold; the deviation weight value is a weighted quantitative index determined based on the identification deviation probability of the identified risk defect type, used to measure the severity of the identification deviation of the identified risk defect type; the preset risk defect type quantity threshold is a critical value used to determine whether the number of identified risk defect types is too large; the preset deviation weight threshold is a critical value used to determine whether the sum of the deviation weight values ​​of all identified risk defect types reaches the critical value that requires adjustment and control management.

[0035] Assuming the identification deviation probability of each quality defect type within the statistical period has been calculated, and a preset deviation probability threshold is set to a certain value, quality defect types with an identification deviation probability greater than this threshold are identified as risk defect types. If the number of identified risk defect types exceeds the preset risk defect type number threshold, it is directly determined that production parameter adjustment and control management is required; if the number does not exceed the threshold, the deviation weight value of each identified risk defect type is further calculated, and the relationship between the sum of the deviation weight values ​​and the preset deviation weight threshold is determined to decide whether adjustment and control management is necessary.

[0036] This step, through the dual judgment of the number of risk defect types and deviation weight values, is significant in that it comprehensively assesses the risk of model identification deviation under the current production parameters from two dimensions: the breadth (number of risk defect types identified) and the depth (total deviation weight values). When the identified deviation problem is relatively common or serious, it triggers adjustment and control management to ensure the accuracy and pertinence of adjustment and control management decisions.

[0037] It should be noted that when there are no quality defect types with a probability of identification deviation greater than the preset deviation probability threshold, it is determined that no adjustment or control management of production parameters is required.

[0038] Assuming that the identification deviation probability of all quality defect types within the statistical period is not greater than the preset deviation probability threshold, it indicates that the image model meets the requirements for the identification accuracy of each quality defect type under the current production parameters. At this time, it is determined that no adjustment and control management of production parameters is required, that is, when the equipment parameters change, they are maintained within the range of the equipment parameters corresponding to the baseline parameters.

[0039] The significance of this situation is that when the overall risk of model identification deviation is controllable, unnecessary adjustments to production parameters can be avoided, the stability of production line operation can be maintained, and the loss of production efficiency caused by frequent adjustments to production parameters can be reduced.

[0040] Additionally, it is understandable that when there are quality defect types with an identification deviation probability greater than a preset deviation probability threshold, these quality defect types with an identification deviation probability greater than the preset deviation probability threshold will be identified as risk defect types, including the following: Case 1: If the number of identified risk defect types is greater than the preset risk defect type number threshold, then it is determined that production parameter adjustment and control processing is required.

[0041] The preset threshold for the number of risk and defect types refers to a critical value used to determine whether the number of identified risk and defect types has reached the level that requires immediate adjustment and control. When the number of identified risk and defect types exceeds this threshold, it indicates that the model identification deviation is relatively common and immediate intervention is needed to adjust and control the production parameters.

[0042] Assuming we statistically identify the number of risk defect types, and set a preset threshold for the number of risk defect types, if the number of identified risk defect types exceeds this threshold, it indicates that the identification accuracy of many quality defect types is insufficient. Therefore, it is necessary to adjust production parameters to improve the model's identification performance, thus directly determining that production parameter adjustment and control are required.

[0043] The significance of directly judging the number of risk defect types in this situation lies in the fact that when the model identifies a large number of quality defect types involved in the deviation problem, it can quickly trigger adjustment and control management decisions, thereby enabling the model to identify and process more quality defect types and ensure the reliability of product quality testing in a timely manner.

[0044] Case 2: If the number of identified risk defect types is not greater than the preset threshold for the number of risk defect types, the deviation weight value of the identified risk defect type is determined based on the identification deviation probability of different identified risk defect types. It is then determined whether the sum of the deviation weight values ​​of different identified risk defect types is greater than the preset deviation weight threshold. If so, it is determined that production parameter adjustment and control processing is required; otherwise, it is determined that production parameter adjustment and control processing is not required.

[0045] The deviation weight value is determined based on the identification deviation probability of the identified risk defect type. The higher the identification deviation probability, the larger the deviation weight value. The preset deviation weight threshold is used to determine whether the sum of the deviation weight values ​​of each identified risk defect type reaches the critical value that requires adjustment and control management.

[0046] Assuming the number of identified risk defect types is no greater than a preset threshold, the deviation weight value for each identified risk defect type is calculated based on its identification deviation probability, and the sum of these deviation weight values ​​is obtained. Let the preset deviation weight threshold be a certain value. If the sum of the deviation weight values ​​is greater than this threshold, it indicates that although the number of identified risk defect types is small, the identification deviation of these types is significant, and production parameter adjustment and control are still required. If the sum of the deviation weight values ​​is no greater than this threshold, then no adjustment and control is required.

[0047] The significance of judging the sum of deviation weight values ​​in this case is that when the number of identified risk defect types is small, it is necessary to further assess the severity of the identification deviation of these types, supplement the judgment of the necessity of adjustment and control management from the perspective of deviation depth, and avoid missing serious identification deviation problems due to simply relying on quantity judgment.

[0048] This embodiment achieves intelligent judgment of the necessity of production parameter adjustment and control management through comprehensive judgment of S11 to S13 and multiple situations. Its core value is reflected in three aspects: First, by statistically analyzing the number of identified deviations and calculating the probability of identified deviations, the model identifies deviations as comparable probability indicators; second, by screening the types of identified risk defects, it identifies quality defect types with insufficient identification accuracy; third, by making dual judgments on the number of identified risk defect types and deviation weight values, it comprehensively evaluates the risk of identified deviations from both breadth and depth dimensions, triggering adjustment and control management decisions when the identified deviation problem is relatively common or serious, and maintaining production stability when the identified deviation problem is controllable, thus achieving refined management of adjustment and control management decisions.

[0049] A yoga apparel manufacturer deployed image models on its production line to conduct real-time quality inspections of its products. The models cover eight types of quality defects: Type A (loose fabric fibers), Type B (abnormal color difference), Type C (seam misalignment), Type D (inadequate elasticity), Type E (pattern printing misalignment), Type F (neckline size deviation), Type G (uneven waist elastic band), and Type H (uneven fabric thickness). During the statistical period of the most recent month, the number of identification and processing times and the number of identification deviations for each type of quality defect are as follows: Type A: 200 identification and processing times, 16 identification deviations; Type B: 180 identification and processing times, 9 identification deviations; Type C: 220 identification and processing times, 33 identification deviations; Type D: 150 identification and processing times, 21 identification deviations; Type E: 160 identification and processing times, 8 identification deviations; Type F: 190 identification and processing times, 6 identification deviations; Type G: 170 identification and processing times, 27 identification deviations; Type H: 140 identification and processing times, 5 identification deviations.

[0050] In S11, the number of identification deviations for each quality defect type was counted, and the results are as follows: Type A: 16 times, Type B: 9 times, Type C: 33 times, Type D: 21 times, Type E: 8 times, Type F: 6 times, Type G: 27 times, and Type H: 5 times.

[0051] In S12, calculate the identification bias probability for each quality defect type: Type A is 16÷200=0.080, Type B is 9÷180=0.050, Type C is 33÷220=0.150, Type D is 21÷150=0.140, Type E is 8÷160=0.050, Type F is 6÷190≈0.032, Type G is 27÷170≈0.159, and Type H is 5÷140≈0.036.

[0052] In S13, the preset deviation probability threshold is set to 0.10. Quality defect types with a deviation probability greater than 0.10 are identified as risk defect types: type C (0.150), type D (0.140), and type G (0.159). The number of risk defect types identified is 3.

[0053] Set the preset threshold for the number of risk defect types to 4. Determine if 3 is greater than 4: if 3 is not greater than 4, proceed to case 2.

[0054] In scenario 2, the deviation weight value is determined based on the identification deviation probability of each identified risk defect type. Let the deviation weight value be calculated as: Deviation weight value = Identification deviation probability ÷ Preset deviation probability threshold. Then: Deviation weight value for type C = 0.150 ÷ ​​0.10 = 1.50, Deviation weight value for type D = 0.140 ÷ 0.10 = 1.40, Deviation weight value for type G = 0.159 ÷ 0.10 = 1.59.

[0055] The total weighted deviation values ​​are: 1.50 + 1.40 + 1.59 = 4.49.

[0056] Set the preset deviation weight threshold to 4.00, and determine whether 4.49 is greater than 4.00: 4.49 is greater than 4.00, therefore it is determined that the production parameters need to be adjusted and controlled, and proceed to S2 to determine the change adjustment scheme.

[0057] S2 determines the changes in the equipment parameters of different production equipment based on the monitoring data of the equipment parameters of the production equipment, and determines the adjustment plan for the production equipment based on the changes in the equipment parameters of different production equipment. The production equipment refers to various processing equipment on the yoga wear production line, including knitting machines, heat-setting machines, printing machines, sewing machines, etc.; the equipment parameters refer to key parameters controlling the operating status of the production equipment, including equipment temperature, speed, tension, pressure, etc.; the variation is determined based on the time period when the variation rate of the equipment parameters of the production equipment meets the requirements, specifically, the time period when the equipment parameters are not in the equipment parameter range corresponding to the benchmark equipment parameters, and the absolute value of the minimum deviation rate of the average value of the equipment parameters from the endpoints of the equipment parameter range corresponding to the benchmark equipment parameters is greater than a preset deviation rate threshold; the benchmark equipment parameters are determined based on the optimal equipment parameter range corresponding to the production equipment determined by historical quality inspection data; the qualified range refers to the equipment parameter range in which the equipment parameters of the production equipment must operate.

[0058] Assume a yoga apparel production line has multiple production machines, and the parameters of each machine are monitored in real time. By comparing the relationship between the machine parameters at different times and the baseline machine parameter range, the variation matching period for each production machine is identified. Preset threshold values ​​are defined as follows: a preset weight threshold, a preset variation weight threshold, a preset baseline ratio threshold, a preset matching coefficient threshold, and first and second duration thresholds. An adjustment plan is determined based on the distribution of the variation matching period.

[0059] This step establishes a technical path from monitoring equipment parameter data to determining adjustment schemes. Its significance lies in assessing the overall parameter stability of production equipment based on the changes in equipment parameters of different production equipment, and then determining under what conditions adjustments should be made when equipment parameters change. The more stable the overall equipment parameters of the production equipment, the more aggressive the adjustment scheme can be. This not only ensures the stability of equipment parameters, but also enables the production equipment to operate under more equipment parameter conditions, providing the model with more opportunities to verify and identify quality defect types.

[0060] Furthermore, the variation of the equipment parameters of the production equipment is determined based on the time period during which the variation rate of the equipment parameters of the production equipment meets the requirements.

[0061] Furthermore, the time period in which the variation rate of the equipment parameters of the production equipment meets the requirements is the time period in which the equipment parameters of the production equipment are not in the equipment parameter range corresponding to the reference equipment parameters, and the absolute value of the minimum value of the deviation rate between the average value of the equipment parameters and the endpoint of the equipment parameter range corresponding to the reference equipment parameters of the production equipment is greater than the preset deviation rate threshold.

[0062] Specifically, the baseline equipment parameters of the production equipment are determined based on the recommended values ​​of the baseline equipment parameters of the production equipment, that is, the optimal equipment parameter range corresponding to the production equipment determined based on historical quality inspection data, while the qualified range is the equipment parameter range in which the production equipment must operate.

[0063] It should be noted that, as Figure 3 As shown, the method for determining the adjustment plan for the production equipment is as follows: In this embodiment, based on the changes in the equipment parameters of different production equipment, the distribution data of the time periods during which the equipment parameters of different production equipment changed are determined. Using the distribution data of the time periods during which the equipment parameters of different production equipment changed, the stability of the equipment parameters of different production equipment is determined. The stability is used to determine the adjustment scheme for all production equipment. That is, the more stable the overall equipment parameters of the production equipment, the more aggressive the adjustment scheme for all production equipment. On the one hand, this ensures the stability of the equipment parameters of the production equipment, and on the other hand, it ensures that the production equipment can operate under more equipment parameters. It also lays the foundation for providing the model with more opportunities to verify and identify quality defect types.

[0064] S21 determines the time period in which the rate of change of the equipment parameters of the production equipment meets the requirements based on the changes in the equipment parameters of different production equipment; The time period in which the rate of change meets the requirement refers to the time period in which the equipment parameters are not within the equipment parameter range corresponding to the benchmark equipment parameters, and the absolute value of the minimum deviation rate between the average value of the equipment parameters and the endpoints of the equipment parameter range is greater than the preset deviation rate threshold, that is, the time period in which the equipment parameters deviate significantly from the benchmark range.

[0065] Suppose that the equipment parameters of a certain production equipment are monitored in real time, and the equipment parameters for each time period are compared with a baseline equipment parameter range. First, time periods in which the equipment parameters are not within the baseline range are selected. Then, the deviation rate between the average equipment parameter value and the endpoint of the baseline range in these time periods is calculated. Time periods in which the absolute value of the minimum deviation rate is greater than a preset deviation rate threshold are retained as the time periods in which the rate of change of the production equipment meets the requirements.

[0066] This step, through dual screening of the rate of change, is significant in identifying periods when equipment parameters not only deviate from the baseline range but also deviate significantly, excluding cases where equipment parameters only slightly deviate from the baseline range, and ensuring that subsequent analysis focuses on the periods of equipment parameter changes that truly affect production quality.

[0067] S22 defines the time period during which the rate of change of the equipment parameters of the production equipment meets the requirements as the change matching time period of the production equipment. The change matching period refers to the period during which the degree of change in equipment parameters meets the requirements after screening by S21. It is used to characterize the historical situation of significant changes in the equipment parameters of the production equipment. It should be noted that during the change matching period and when adjusting the equipment parameters, it is necessary to ensure that the equipment parameters of the production equipment are operating within the qualified range at different times. If they are not within the qualified range, adjustment should be carried out immediately.

[0068] Suppose that after a certain production equipment is screened by S21, there are multiple time periods in which the change rate meets the requirements. These time periods are uniformly marked as the change matching time periods of the production equipment, and the number and duration of the change matching time periods of each production equipment are counted.

[0069] This step, by marking the change matching time period, is significant in providing a standardized data foundation for subsequent calculation of data change weight values ​​and determination of change adjustment schemes. At the same time, it emphasizes that equipment parameters must always remain within the qualified range during any change adjustment process to ensure that production safety and product quality are not affected.

[0070] S23 determines the adjustment plan for the production equipment based on the different time periods of change matching for different production equipment.

[0071] The aforementioned change adjustment scheme specifies the conditions under which equipment parameter adjustments are required when the equipment parameters of production equipment change (i.e., are outside the baseline equipment parameter range). The aforementioned data change weight value refers to the proportion of the duration of the change matching period of a certain production equipment to the total monitoring time, used to quantify the instability of the equipment parameters of that production equipment. The aforementioned screened production equipment refers to production equipment whose data change weight value is above a preset change weight threshold (less than the preset weight threshold). The aforementioned equipment change matching coefficient is a comprehensive index that considers both the proportion of screened production equipment and the data change weight value of each production equipment, used to assess the overall severity of equipment parameter changes in the production equipment.

[0072] Assuming the timeframe for matching changes in each production device is determined, calculate the data change weight value for each production device. Set a preset weight threshold, a preset change weight threshold (less than the preset weight threshold), a preset baseline proportion threshold, and a preset matching coefficient threshold. Based on whether any production device has a data change weight value greater than the preset weight threshold, and considering the proportion of production devices and the size of the device change matching coefficient, determine the type of adjustment scheme.

[0073] This step, through multi-layered judgments on data change weight values, the proportion of production equipment selected, and the matching coefficient of equipment change, is significant in that it dynamically selects the strictness of the change adjustment scheme based on the overall stability of the production equipment parameters. When the overall parameters of the production equipment are stable, an aggressive scheme is adopted, and when the overall parameters are unstable, a conservative scheme is adopted. While ensuring the stability of equipment parameters, it enables the equipment to operate under more parameter conditions, thus accumulating more recognition training data for the model.

[0074] It is understandable that if there is no matching period for the change of different production equipment, the equipment parameters of different production equipment are relatively stable. Therefore, the change adjustment scheme of the production equipment is determined to be a screening adjustment scheme. That is, when the equipment parameters change, only when the duration of the change of the equipment parameters, that is, the duration of the change that is not within the equipment parameter range corresponding to the baseline equipment parameters, is at the first duration threshold, is it necessary to adjust the equipment parameters.

[0075] The aforementioned screening and adjustment scheme refers to a scheme that sets a relatively long duration threshold (first duration threshold) as the adjustment trigger condition when the equipment parameters change. That is, adjustment is only performed when the duration of the equipment parameters deviating from the reference range reaches the first duration threshold. The first duration threshold refers to the critical value of the duration of the equipment parameters deviating from the reference range that triggers the equipment parameter adjustment process.

[0076] Assuming that no matching periods of change exist for any production equipment within the statistical period, it indicates that the overall operation of the equipment parameters of each production equipment is stable, and the deviations from the baseline range are relatively small. In this case, the change adjustment scheme is determined as the screening adjustment scheme, and a relatively long first time threshold is used as the adjustment trigger condition. This allows the equipment parameters to deviate from the baseline range within a certain time period without adjustment, thereby enabling the equipment to operate under a wider range of parameter conditions and providing the model with more opportunities for identification and processing.

[0077] The significance of determining the adjustment scheme in this case is that when the overall parameters of the production equipment are stable, the most aggressive adjustment scheme is adopted to maximize the range of changes in the equipment parameters, accumulate as much identification training data as possible under different parameter conditions for the model, and accelerate the iterative optimization process of the model.

[0078] Additionally, it should be noted that if there are changes in the production equipment during the matching period, the following should be included: S211 uses the change matching time period data of different production equipment to determine the duration ratio of the change matching time period of different production equipment. The duration ratio of the change matching time period of the production equipment is used as the data change weight value. It is determined whether there is a production equipment whose data change weight value is greater than the preset weight threshold. If so, the change adjustment scheme of the production equipment is determined to be that when the equipment parameters change, the equipment parameter adjustment is only required when the duration of the change of the equipment parameters, that is, the duration of the change that is not within the equipment parameter range corresponding to the baseline equipment parameters, is the second duration threshold (where the second duration threshold is less than the first duration threshold). Otherwise, proceed to step S212.

[0079] The data change weight value refers to the proportion of the total duration of the change matching period of a certain production equipment to the total monitoring duration; the preset weight threshold refers to the critical value used to judge whether the data change weight value is too high; the second duration threshold refers to the adjustment trigger duration threshold that is less than the first duration threshold, which represents a more conservative adjustment trigger condition compared to the screening adjustment scheme.

[0080] Assuming there are production equipment with matching change periods, calculate the data change weight value for each production equipment. Assume a preset weight threshold is a certain value. If the data change weight values ​​of all production equipment are not greater than the preset weight threshold, it indicates that although there are matching change periods, the overall degree of change is not severe. In this case, proceed to the next step. Otherwise, if the degree of change is more severe, determine that the change adjustment scheme uses a second duration threshold as the adjustment trigger condition. That is, adjustment is only performed when the duration of equipment parameters deviating from the baseline range reaches the second duration threshold.

[0081] This step compares the data change weight value with the preset weight threshold. Its significance lies in the fact that when there is a change matching period in the production equipment but the overall change degree is relatively serious, a slightly more conservative adjustment scheme (the second duration threshold replaces the first duration threshold) is adopted. This ensures that the equipment parameters do not deviate significantly from the baseline range for a long time, while still providing the model with a certain number of identification and processing opportunities under different parameter conditions.

[0082] S212 Based on the data change weight values ​​of different production equipment, determine the production equipment whose data change weight values ​​are above the preset change weight threshold (less than the preset weight threshold) and use them as screening production equipment. Determine whether the proportion of screening production equipment in the production equipment is greater than the preset benchmark proportion threshold. If yes, proceed to step S213. If no, determine that the change adjustment scheme for the production equipment is that when the equipment parameters change, only when the duration of the change in the equipment parameters, that is, the duration of the change that is not within the equipment parameter range corresponding to the benchmark equipment parameters, is the first duration threshold, is it necessary to perform equipment parameter adjustment processing. The preset change weight threshold refers to the weight threshold used to identify production equipment with relatively frequent changes in equipment parameters, and is less than the preset weight threshold; the screened production equipment refers to production equipment with a data change weight value above the preset change weight threshold, indicating that the equipment parameters of the production equipment have relatively frequent changes; the preset benchmark ratio threshold refers to the threshold used to determine whether the proportion of the screened production equipment in the total production equipment is high.

[0083] Assuming there are production equipment whose data change weight values ​​exceed a preset weight threshold, further screening is performed on production equipment whose data change weight values ​​are also above the preset change weight threshold. The proportion of the screened production equipment in all production equipment is then calculated. Let the preset baseline proportion threshold be a certain value. If the proportion of the screened production equipment is not greater than this threshold, it indicates that the number of production equipment with frequently changing parameters is relatively small. In this case, the adjustment scheme is determined using a first duration threshold as the adjustment trigger condition, i.e., a screening adjustment scheme is adopted, maintaining a relatively aggressive adjustment strategy.

[0084] The significance of this step, which involves judging the proportion of production equipment, lies in the fact that when the proportion of production equipment with more frequent changes in equipment parameters is low, the overall production line parameters remain relatively stable, and aggressive adjustment schemes can still be adopted to accumulate more recognition training data for the model.

[0085] S213 Based on the proportion of screened production equipment in the production equipment and the data change weight values ​​of different production equipment, determine the equipment change matching coefficient, and determine whether the equipment change matching coefficient is greater than the preset matching coefficient threshold. If so, determine that the production equipment change adjustment scheme is that when the equipment parameters change, equipment parameter adjustment is only required when the duration of the equipment parameter change, i.e., the duration of the change that is not within the equipment parameter range corresponding to the benchmark equipment parameter, is greater than the second duration threshold, or when there is a period of equipment parameter adjustment processing within the most recent preset duration. If not, determine that the production equipment change adjustment scheme is that when the equipment parameters change, equipment parameter adjustment is only required when the duration of the equipment parameter change, i.e., the duration of the change that is not within the equipment parameter range corresponding to the benchmark equipment parameter, is the second duration threshold.

[0086] The equipment change matching coefficient refers to a comprehensive evaluation index that takes into account the proportion of production equipment selected and the weight value of data changes of each production equipment. The preset matching coefficient threshold refers to the critical value used to judge whether the equipment change matching coefficient is too high. When the change adjustment plan is triggered when the duration is above the second duration threshold or when there is no adjustment processing period in the most recent preset duration, the adjustment triggering conditions are more diverse than the plan that only uses the second duration threshold as a condition. That is, when the equipment parameters deviate from the benchmark range for a certain duration or when no adjustment processing has been carried out recently, a new round of adjustment will be triggered, indicating that the control of equipment parameter changes is more stringent.

[0087] Assuming the proportion of screened production equipment exceeds a preset baseline threshold, the equipment variation matching coefficient is further calculated. Let the preset matching coefficient threshold be a certain value. If the equipment variation matching coefficient is greater than this threshold, it indicates that the overall equipment parameters of the production equipment are significantly affected. In this case, a more stringent adjustment scheme is adopted. The trigger condition is either a duration exceeding the second duration threshold or the existence of an adjustment processing period within the most recent preset duration. That is, if no adjustment processing period exists within the most recent preset duration, adjustment can only be performed when the duration exceeds the second duration threshold. If an adjustment processing period exists, adjustment is performed immediately to ensure the equipment operates within the parameter range corresponding to the baseline equipment parameters. If the equipment variation matching coefficient is not greater than this threshold, it indicates that the overall variation is still controllable, and the second duration threshold remains the sole adjustment trigger condition.

[0088] It should be noted that the adjustment processing period is the period after the second duration threshold is reached, and when the adjustment processing is performed, the period when the device is running outside the device parameter range corresponding to the device parameters and reaches the second duration threshold is taken as the adjustment processing period.

[0089] This step, through the determination of the equipment change matching coefficient, is significant in that it dynamically selects the most suitable change adjustment scheme based on the severity of the overall equipment parameter changes in the production equipment. When the changes are severe, a strict scheme with multiple trigger conditions is adopted to ensure the stability of equipment parameters, while when the changes are controllable, a scheme with a single trigger condition is adopted to moderately allow changes in equipment parameters, thereby achieving refined management of equipment parameter adjustments.

[0090] This embodiment achieves intelligent determination of production equipment change adjustment schemes through comprehensive judgment of S21 to S23 and multiple situations. Its core value is reflected in four aspects: First, by double screening the time period when the change rate meets the requirements, it accurately identifies the time period when the equipment parameters deviate from the benchmark range and the magnitude is significant; second, by calculating the data change weight value, it quantifies the degree of instability of equipment parameters of each production equipment; third, by screening the proportion of production equipment and the matching coefficient of equipment change, it evaluates the overall parameter stability of production equipment; fourth, by flexibly switching between the screening adjustment scheme, the scheme based on the first time threshold, the scheme based on the second time threshold, and the multi-trigger condition scheme, it achieves a dynamic balance between ensuring the stability of equipment parameters and accumulating model training data.

[0091] Continuing with the example in S1, it has been determined that adjustments and control of production parameters are required. There are a total of 12 production machines (machines 1 to 12) on the yoga apparel production line, with a total monitoring duration of 30 days.

[0092] In S21, a variation rate analysis is performed on the equipment parameters of each production device. A preset deviation rate threshold of 5% is set, and periods in which the absolute value of the minimum deviation rate of each production device's equipment parameters is outside the baseline range are counted. Assume the variation matching periods for devices 1 to 12 are as follows: Device 1 has 3 variation matching periods (total duration 18 hours), Device 2 has 2 periods (total duration 12 hours), Device 3 has 5 periods (total duration 30 hours), Device 4 has 0 periods (no variation matching periods), Device 5 has 4 periods (total duration 24 hours), Device 6 has 0 periods (no variation matching periods), Device 7 has 1 period (total duration 6 hours), Device 8 has 0 periods (no variation matching periods), Device 9 has 3 periods (total duration 20 hours), Device 10 has 0 periods (no variation matching periods), Device 11 has 2 periods (total duration 10 hours), and Device 12 has 0 periods (no variation matching periods).

[0093] In S22, the time periods during which the above-mentioned change rates meet the requirements are taken as the change matching time periods for the corresponding production equipment. Equipment 4, Equipment 6, Equipment 8, Equipment 10, and Equipment 12 have no change matching time periods, while the remaining 7 equipment have change matching time periods.

[0094] In S23, proceed to S211 to calculate the data change weight value of each production equipment (total duration of change matching period ÷ 720 hours): Equipment 1 is 18 ÷ 720 ≈ 0.025, Equipment 2 is 12 ÷ 720 ≈ 0.017, Equipment 3 is 30 ÷ 720 ≈ 0.042, Equipment 4 is 0, Equipment 5 is 24 ÷ 720 ≈ 0.033, Equipment 6 is 0, Equipment 7 is 6 ÷ 720 ≈ 0.008, Equipment 8 is 0, Equipment 9 is 20 ÷ 720 ≈ 0.028, Equipment 10 is 0, Equipment 11 is 10 ÷ 720 ≈ 0.014, Equipment 12 is 0.

[0095] Set the preset weight threshold to 0.040. Determine if there is any production equipment with a data change weight value greater than 0.040: The data change weight value of equipment 3 is 0.042, which is greater than 0.040. Therefore, there is a production equipment with a data change weight value greater than the preset weight threshold, and proceed to S212.

[0096] In S212, the preset change weight threshold is set to 0.020 (less than the preset weight threshold of 0.040). The production equipment with a data change weight value of 0.020 or higher is filtered as follows: Equipment 1 (0.025), Equipment 3 (0.042), Equipment 5 (0.033), and Equipment 9 (0.028). A total of 4 production equipment are filtered.

[0097] The proportion of production equipment to be screened is approximately 4 ÷ 12 ≈ 0.333. Assuming a preset baseline proportion threshold of 0.30, determine if 0.333 is greater than 0.30: if 0.333 is greater than 0.30, proceed to step S213.

[0098] In S213, calculate the equipment change matching coefficient. Let the equipment change matching coefficient = proportion of screened production equipment × average value of change weight of screened production equipment data ÷ preset change weight threshold. The average value of change weight of screened production equipment data = (0.025 + 0.042 + 0.033 + 0.028) ÷ 4 = 0.032. The equipment change matching coefficient = 0.333 × 0.032 ÷ 0.020 (0.02 is the scaling factor) = 0.333 × 1.60 = 0.533.

[0099] Set the preset matching coefficient threshold to 0.50, and determine whether 0.533 is greater than 0.50: 0.533 is greater than 0.50, therefore the adjustment scheme for the production equipment is determined as follows: When the equipment parameters change, the equipment parameters need to be adjusted only if the duration of the deviation of the equipment parameters from the reference range exceeds the second duration threshold (set to be 10 minutes), or if there is a period for adjusting the equipment parameters within the most recent preset duration (set to be 24 hours). That is, if there is a period for adjusting the equipment parameters within the most recent preset period, there is already a period in which the equipment parameters have operational deviations, so the adjustment is performed directly. In other cases, the adjustment is performed only after the duration of the deviation of the equipment parameters from the reference range exceeds the second duration threshold (set to be 10 minutes).

[0100] If the equipment variation matching coefficient is 0.45, which is not greater than the preset matching coefficient threshold of 0.50, then the variation adjustment plan is determined as follows: equipment parameter adjustment is only required when the duration of the equipment parameter deviation from the baseline range is 10 minutes below the second duration threshold.

[0101] S3 utilizes the aforementioned adjustment scheme to adjust the equipment parameters of the production equipment when changes occur. Based on the changes in the identification processing data of the image model under different quality defect types, and the number of identification deviations under different quality defect types, when it is determined that active adjustment targets in the production equipment need to be identified, the active adjustment targets in the production equipment are determined based on the changes in the equipment parameters of the production equipment and the adjustment scheme.

[0102] The proactive adjustment target refers to production equipment that does not have a period of equipment parameter change within the recent preset time period and requires proactive equipment parameter adjustment; the increase in the number of identification processing times refers to the increase in the number of identification processing times for each quality defect type compared to before the adjustment, after the equipment parameters are adjusted according to the change adjustment plan; the newly added deviation defect type refers to the quality defect type whose increase in the number of identification processing times is less than a preset threshold; the risk type ratio refers to the proportion of newly added deviation defect types that belong to the identified risk defect type; the identified risk value refers to the risk quantification index determined comprehensively based on the number of identified risk defect types and the risk type ratio.

[0103] Assuming that equipment parameters are adjusted according to the change and adjustment plan determined in S2, the number of new identification and processing times for each quality defect type is statistically analyzed. Let there be a preset threshold for the number of new defects, a preset threshold for the number of new defect types, a preset threshold for the risk ratio, a preset risk threshold, a preset threshold for the proportion of changed production equipment, a preset threshold for the number of changed time periods, a preset threshold for the matching factor, a preset ratio, and a preset target quantity. Based on the number of new deviation defect types and the proportion of risk types, it is determined whether proactive adjustment target identification and processing is necessary.

[0104] This step establishes a technical path from the implementation effect of the change and adjustment plan to the determination of the proactive adjustment target. Its significance lies in evaluating the model verification processing efficiency of the change and adjustment plan under different quality defect types, identifying defect types that have insufficient increase in the number of identification and processing steps and overlap with the identified risk defect types, and proactively identifying and determining the production equipment that needs parameter adjustment as the proactive adjustment target when the current change and adjustment plan cannot meet the model verification requirements, thereby further improving the reliability and efficiency of the model verification processing.

[0105] Specifically, determining the need for active adjustment target identification processing in the production equipment includes: In this embodiment, based on the increase in the number of identification processing times under different quality defect types after utilizing the change and adjustment scheme, and the degree of overlap between the quality defect types with identification deviation probabilities greater than a preset deviation probability threshold (i.e., the identification risk defect types), the reliability and efficiency of the model verification process under the current change and adjustment scheme are determined. That is, the fewer the number of identification processing times under different quality defect types, and the higher the degree of overlap between the quality defect types with fewer new processing times and the quality defect types with identification deviation probabilities greater than a preset deviation probability threshold, the lower the reliability and efficiency of the model verification process under the current change and adjustment scheme. The reliability and efficiency of the model verification process under the current change and adjustment scheme are then used to determine whether it is necessary to perform active adjustment target identification processing in the production equipment, i.e., to determine the production equipment that actively performs equipment parameter adjustment processing, further improving the reliability and efficiency of the model verification process.

[0106] S31 determines the number of new identification and processing times for different quality defect types after the image model is adjusted according to the production equipment change plan based on the changes in the identification and processing data of the image model under different quality defect types. The change in the identification and processing data refers to the change in the number of identification and processing times for each quality defect type before and after the equipment parameters are adjusted according to the change and adjustment plan; the increase in the number of identification and processing times refers to the number of new identification and processing times after the change and adjustment plan is implemented, which is used to evaluate the improvement effect of the change and adjustment plan on the model's identification and processing opportunities.

[0107] Assuming that the equipment parameters are adjusted according to the change and adjustment plan, the number of times each quality defect type is identified and processed within a certain period of time is counted. Compared with before the adjustment, the increase in the number of times each quality defect type is identified and processed is calculated, forming a dataset of the increase in the number of each quality defect type.

[0108] This step, by statistically analyzing the increase in the number of processing attempts, is significant in quantifying the improvement effect of the adjustment scheme on the identification and processing opportunities of various quality defect types in the image model. It identifies quality defect types with insufficient increase in the number of processing attempts, providing a data basis for subsequent identification of new deviation defect types and determination of the necessity of proactive adjustment.

[0109] S32 uses the newly added quantity to determine the newly added deviation defect type in the quality defect type, and determines the degree of overlap between the newly added deviation defect type and the identified risk defect type based on the number of identified deviations under different quality defect types.

[0110] The newly added deviation defect type is a quality defect type whose newly added quantity is less than the preset newly added quantity threshold, that is, a quality defect type whose number of identification processing times is insufficient after the change and adjustment plan is implemented; the degree of overlap refers to the proportion and quantity of newly added deviation defect types that belong to the identification risk defect type (that is, the quality defect type whose identification deviation probability determined by S1 is greater than the preset deviation probability threshold).

[0111] Assuming the number of new identification and processing attempts for each quality defect type has been counted, and a preset threshold for the number of new attempts is set, quality defect types with a number of new attempts less than this threshold are considered as newly identified deviation defect types. Further comparison is made between these newly identified deviation defect types and the identified risk defect types determined in S1, and the number and proportion of overlapping defect types are calculated.

[0112] This step, through the analysis of the degree of overlap between the newly added deviation defect types and the identified risk defect types, is significant in that it assesses the validation coverage of the current change and adjustment plan on the quality defect types with the highest model identification risk. The lower the degree of overlap, the lower the support of the change and adjustment plan for the identification and processing of the quality defect types that most need validation, and the higher the necessity for proactive adjustment.

[0113] Based on the newly added deviation defect type and the degree of overlap between the newly added deviation defect type and the identified risk defect type, S33 determines whether it is necessary to perform the identification processing of the active adjustment target in the production equipment.

[0114] The preset threshold for the number of newly added defect types refers to a critical value used to determine whether the number of newly added deviation defect types is too large; the preset risk ratio threshold refers to a critical value used to determine whether the risk type ratio is too high; the risk type ratio refers to the proportion of newly added deviation defect types that belong to the identified risk defect types to the total number of newly added deviation defect types.

[0115] Assuming the number of newly added deviation defect types is already determined, a preset threshold for the number of newly added defect types is set to a certain value. If the number of newly added deviation defect types exceeds this threshold, it is directly determined that proactive adjustment is required; if the number does not exceed this threshold, the degree of overlap is further evaluated, and a decision on whether proactive adjustment is needed is made based on the proportion of risk types and the identified risk value.

[0116] This step, through a comprehensive judgment of the number of newly added deviation defect types and the degree of overlap, is significant in that it comprehensively assesses the necessity of proactive adjustment from two dimensions: the breadth of insufficient identification and processing (the number of newly added deviation defect types) and the degree of correlation with the identified risk defect types (the proportion of risk types and the identified risk value). This avoids unnecessary proactive adjustment while ensuring that the identified quality defect types with high deviation risk have sufficient opportunities for verification and identification.

[0117] Furthermore, if the number of newly added deviation defect types is greater than a preset threshold for the number of newly added defect types, it is determined that active adjustment target identification processing in the production equipment is required.

[0118] If the number of newly added deviation defect types exceeds the preset threshold for the number of newly added defect types, it indicates that under the current adjustment plan, there are too many quality defect types that cannot be identified and processed in sufficient numbers, and the adjustment plan has failed to provide the model with enough opportunities for identification and processing. In this case, it is determined that proactive adjustment of the target identification process is needed. This can be achieved by proactively adjusting the equipment parameters of some production equipment to increase the identification and processing opportunities for each quality defect type.

[0119] The significance of this situation lies in the fact that when the overall effect of the change and adjustment plan is insufficient, it can quickly trigger proactive adjustment decisions, thus avoiding the problem of insufficient opportunities to identify and process large-scale issues from further affecting the iterative optimization process of the model.

[0120] Additionally, it is understood that if the number of newly added deviation defect types is not greater than a preset threshold for the number of newly added defect types, the following situations also apply: Scenario 1: If, based on the degree of overlap between the newly added deviation defect type and the identified risk defect type, it is determined that none of the newly added deviation defect types belong to the identified risk defect types, then it is determined that no active adjustment target identification processing is required in the production equipment.

[0121] Assuming the number of newly added deviation defect types does not exceed the preset threshold for the number of newly added defect types, and a comparison between the newly added deviation defect types and the identified risk defect types reveals no overlap, it indicates that the quality defect types with insufficient identification processing times do not belong to the types with high identification deviation risk. The current adjustment plan provides sufficient verification coverage for the most critical identified risk defect types. Therefore, it is determined that no active adjustment target identification processing is needed.

[0122] The significance of this situation is that when the types of quality defects for which the number of identification processes is insufficient are all of low risk of identification deviation, the current adjustment plan can meet the most critical verification requirements without the need for additional proactive adjustments, thus avoiding unnecessary intervention in the production line.

[0123] Scenario 2: If the newly added deviation defect types include risk-identifying defect types, the risk type ratio is determined based on the proportion of the number of risk-identifying defect types among the newly added deviation defect types. It is then determined whether the risk type ratio is greater than a preset risk ratio threshold. If yes, it is determined that active adjustment target identification processing in the production equipment is required; otherwise, proceed to the next step. Based on the number of risk-identifying defect types and the risk type ratio, the identification risk value is determined. It is then determined whether the identification risk value is greater than a preset risk threshold. If yes, it is determined that active adjustment target identification processing in the production equipment is required; otherwise, it is determined that active adjustment target identification processing in the production equipment is not required.

[0124] The risk type ratio refers to the proportion of newly added deviation defect types that belong to the identified risk defect type to the total number of newly added deviation defect types; the identified risk value is determined based on the number of identified risk defect types and the risk type ratio, and the value ranges from 0 to 1. The more identified risk defect types and the higher the risk type ratio, the greater the identified risk value.

[0125] Assuming that some of the newly added deviation defect types include risky defect types, the proportion of risky types is calculated. A preset risk proportion threshold is set. If the risk type proportion is greater than this threshold, it indicates that most of the newly added deviation defect types are of high risk and require proactive adjustment. If the risk type proportion is not greater than this threshold, the identification risk value is further calculated based on the number of risky defect types and the risk type proportion, and compared with the preset risk threshold to comprehensively determine whether proactive adjustment of the target identification process is necessary.

[0126] This approach, through the dual judgment of risk type ratio and risk value identification, is significant because when a new deviation defect type contains an identified risk defect type, the necessity of proactive adjustment is assessed from two dimensions: overlap ratio and overall risk. Proactive adjustment is triggered when the overlap ratio is high or the overall risk is large, and the current plan is maintained when both the overlap ratio and overall risk are controllable, thus achieving refined management of proactive adjustment decisions.

[0127] It is understood that the more types of risk defects identified, the higher the proportion of risk types, and the greater the identified risk value, specifically, its value ranges from 0 to 1.

[0128] It is understood that the method for determining the active adjustment target in the production equipment is as follows: In this embodiment, based on the change data of the equipment parameters of the production equipment and the production equipment with periods of equipment parameter change, the change of equipment parameters of the production equipment under the current change adjustment scheme is determined. That is, the more production equipment with parameter change and the more periods of equipment parameter change of the production equipment with parameter change, the higher the reliability of the change. In other words, the more opportunities the model has to identify and process different types of quality defects, and the higher the verification reliability is. Using the verification reliability, it is determined which production equipment should be used as the active adjustment target. That is, when there are no periods of equipment parameter change in the recent preset time, the production equipment whose equipment parameters are actively adjusted is selected. This also lays the foundation for further improving the comprehensiveness and reliability of the model's verification processing.

[0129] S41 Based on the change data of the equipment parameters of the production equipment, determine the time period in which the equipment parameters of the production equipment are not within the equipment parameter range corresponding to the benchmark equipment parameters, and take the time period in which the equipment parameters of the production equipment are not within the equipment parameter range corresponding to the benchmark equipment parameters as the equipment parameter change period of the production equipment. The equipment parameter change period refers to the period during which the equipment parameters of the production equipment are not within the baseline equipment parameter range. It is used to identify the period during which the equipment parameters of the production equipment actually change within the historical monitoring period.

[0130] Assuming we obtain the equipment parameter change data for each production equipment during the most recent monitoring period, we identify the periods when the equipment parameters are not within the baseline range, mark these periods as the equipment parameter change periods for each production equipment, and count the number of equipment parameter change periods for each production equipment.

[0131] This step, by identifying the time periods of equipment parameter changes, is significant in distinguishing between production equipment whose parameters have changed (with a change period) and production equipment whose parameters have not changed (without a change period), thus providing basic data for subsequent identification of production equipment with parameter changes and determination of proactive adjustment targets.

[0132] S42 defines production equipment that experiences periods of parameter variation as production equipment with parameter variation. The term "parameter-changing production equipment" refers to production equipment whose parameters have changed at least once during the monitoring period. These equipment have already experienced deviations in their parameters, and the model has obtained a certain number of identification and processing opportunities under different parameter conditions.

[0133] Suppose we count the number of time periods during which the equipment parameters of each production equipment change. Production equipment that has at least one time period of change is marked as production equipment with parameter changes. We then count the number of production equipment with parameter changes and their proportion among all production equipment.

[0134] This step, through the identification of production equipment with varying parameters, is significant in distinguishing which production equipment has provided the model with opportunities for identification and processing under different parameter conditions, and which production equipment has not yet provided such opportunities to the model, thus providing a basis for subsequent determination of production equipment that needs to be actively adjusted.

[0135] S43 determines the active adjustment target in the production equipment based on the parameter change production equipment and the time period of parameter change of the production equipment, and in conjunction with the change adjustment plan.

[0136] The variation matching factor refers to a comprehensive evaluation index that takes into account the proportion of the total number of production equipment with drastic changes and production equipment with parameter changes in all production equipment; the production equipment with drastic changes refers to production equipment with parameter changes whose number of periods of equipment parameter changes is greater than a preset threshold; the preset proportion refers to the proportion of production equipment that is not a production equipment with parameter changes that is randomly selected as an active adjustment target; the preset target quantity refers to the specific number of production equipment that is not a production equipment with parameter changes that is randomly selected as an active adjustment target.

[0137] Assuming the proportion of production equipment with variable parameters among all production equipment is determined, and setting a preset threshold for the proportion of variable production equipment, a preset threshold for the number of variable periods, a preset threshold for the matching factor, a preset proportion, and a preset target quantity, then based on the proportion of production equipment with variable parameters, the situation of production equipment with drastic changes, the change matching factor, and the type of change adjustment scheme, determine which production equipment, which does not belong to the category of production equipment with variable parameters, will be the active adjustment targets.

[0138] This step, through a comprehensive evaluation of the proportion of production equipment with parameter changes and the matching factor, is significant in that it dynamically determines the number of active adjustment targets based on the overall reliability of equipment parameter changes. When the reliability of changes is low, all non-parameter-changing production equipment is actively adjusted, while when the reliability of changes is high, only non-parameter-changing production equipment with a preset proportion or a preset target number is actively adjusted, thereby achieving efficient allocation of active adjustment resources.

[0139] Furthermore, if the proportion of the production equipment with variable parameters in the production equipment is less than a preset threshold for the proportion of variable production equipment, then the active adjustment target in the production equipment is determined to be all production equipment that does not belong to the production equipment with variable parameters.

[0140] If the proportion of production equipment with variable parameters among all production equipment is less than a preset threshold for the proportion of variable production equipment, it indicates that the number of production equipment that has experienced parameter changes is relatively small, and the production line as a whole lacks sufficient opportunities for identification and processing under different parameter conditions. In this case, the active adjustment target is determined to be all production equipment that does not belong to the category of production equipment with variable parameters. Active parameter adjustments are made to these equipment to quickly supplement the opportunities for identification and processing under different parameter conditions.

[0141] The significance of this situation is that when the number of production equipment with changing parameters is small, it is necessary to proactively expand the coverage of equipment parameter changes, including all production equipment that has not changed in the proactive adjustment, and quickly supplement the model's recognition training data.

[0142] It is also understood that if the proportion of the production equipment with the parameter variation in the total production equipment is not less than a preset threshold for the proportion of the production equipment with the parameter variation, the following is included: S431, by varying the equipment parameter change periods of the production equipment with different parameters, determine whether there are any production equipment with parameter change periods whose number of change periods exceeds a preset threshold. If yes, proceed to step S432; otherwise, determine that the active adjustment target among the production equipment is all production equipment that does not belong to the parameter change production equipment.

[0143] The preset threshold for the number of change periods refers to a critical value used to determine whether the equipment parameters of production equipment with changing parameters change frequently. If there are no production equipment with changing parameters whose number of change periods exceeds the preset threshold, it indicates that although the number of production equipment with changing parameters is sufficient, the number of change periods for each piece of equipment is relatively small, and the reliability of the equipment parameter changes is low.

[0144] Assuming the proportion of production equipment with variable parameters is not less than a preset threshold for the proportion of production equipment with variable parameters, the number of time periods for parameter changes for each production equipment with variable parameters is further counted. If the number of time periods for changes for all production equipment with variable parameters is not greater than the preset threshold for the number of time periods for changes, it indicates that the number of changes for each production equipment with variable parameters is relatively small, and the overall reliability of the changes is insufficient. In this case, the active adjustment target is determined to be all production equipment that does not belong to the production equipment with variable parameters, thus expanding the coverage of parameter changes.

[0145] This step, by determining whether drastic changes exist in production equipment, is significant in assessing the sufficiency of changes in production equipment with varying parameters. When the sufficiency of changes is insufficient, all non-parameter-changing production equipment is actively adjusted to ensure that the model obtains sufficient opportunities for identification and processing under different parameter conditions.

[0146] S432 identifies production equipment with parameter changes whose number of time periods of change exceeds a preset threshold as drastically changing production equipment. A change matching factor is determined based on the proportion of the total number of drastically changing and parameter-changing production equipment among the production equipment. It is then determined whether the change matching factor exceeds a preset matching factor threshold. If so, the active adjustment target among the production equipment is identified as a preset proportion of production equipment that does not belong to the parameter-changing production equipment category. That is, a preset proportion of production equipment that does not belong to the parameter-changing production equipment category is randomly selected as the active adjustment target. If not, the process proceeds to step S433.

[0147] The term "drastic change production equipment" refers to production equipment whose parameter change period exceeds a preset threshold; the term "change matching factor" is determined by comprehensively considering the proportion of drastic change production equipment and the total number of parameter change production equipment in all production equipment.

[0148] Assuming there are production equipment undergoing drastic changes, calculate the change matching factor. Set a preset threshold value for the matching factor. If the change matching factor is greater than this threshold, it indicates that the overall reliability of parameter changes in the production equipment is high, and it is not necessary to actively adjust all non-parameter-changing production equipment; only a preset proportion of non-parameter-changing production equipment needs to be actively adjusted. If the change matching factor is not greater than this threshold, proceed to step S433 for further judgment.

[0149] This step, through the judgment of the changing matching factor, is significant in that when the overall reliability of the production equipment is high, the number of actively adjusted targets is reduced, and while maintaining the stability of the production line, the coverage of parameter changes is appropriately increased, thereby achieving efficient allocation of actively adjusted resources.

[0150] S433 determines whether the change adjustment scheme belongs to the screening adjustment scheme. If yes, then the active adjustment target in the production equipment is determined to be a preset proportion of production equipment that does not belong to the parameter change production equipment. That is, among the production equipment that does not belong to the parameter change production equipment, a preset proportion of production equipment is randomly selected as the active adjustment target. If no, then the active adjustment target in the production equipment is determined to be a preset proportion of production equipment that does not belong to the parameter change production equipment or a preset target number of production equipment that does not belong to the parameter change production equipment. That is, among the production equipment that does not belong to the parameter change production equipment, a preset proportion or a preset target number of production equipment is randomly selected as the active adjustment target.

[0151] Randomly select a preset proportion or a preset target number of production equipment as the active adjustment target. Specifically, if the preset proportion of production equipment that is not a production equipment with parameter changes is greater than the preset target number, then select the preset proportion of production equipment that is not a production equipment with parameter changes as the active adjustment target. Otherwise, select the preset target number of production equipment that is not a production equipment with parameter changes as the active adjustment target.

[0152] The screening and adjustment scheme is the most aggressive change and adjustment scheme determined in S2, that is, the scheme that uses the first duration threshold as the adjustment trigger condition; when the change and adjustment scheme belongs to the screening and adjustment scheme, it indicates that the overall parameters of the production equipment are stable, and the preset ratio is used for active adjustment; when the change and adjustment scheme does not belong to the screening and adjustment scheme, it indicates that the stability of the parameters of the production equipment is relatively low, and the preset ratio and preset target quantity can be selected for active adjustment to further reduce the degree of intervention in the production line.

[0153] Assuming the change matching factor is not greater than the preset matching factor threshold, the type of change adjustment scheme is further determined. If the change adjustment scheme is a screening adjustment scheme, it indicates that the delay in the active adjustment of production equipment parameters is relatively high, providing more opportunities. The active adjustment target is determined to be the production equipment with non-parametric changes in the preset proportion. If the change adjustment scheme is not a screening adjustment scheme, the size of the preset proportion number and the preset target number are further compared. The method with the smaller number is selected to determine the active adjustment target, so as to minimize the intervention on the production line while meeting the active adjustment requirements.

[0154] This step, by determining the type of change and adjustment plan, is significant in linking the aggressiveness of the change and adjustment plan with the number of proactive adjustment targets. When the change and adjustment plan is more aggressive, the number of proactive adjustment targets is increased; when the plan is more conservative, the number of proactive adjustment targets is reduced, thereby achieving overall coordination and consistency between the determination of the change and adjustment plan and the proactive adjustment targets.

[0155] Continuing with the S2 implementation, the change adjustment scheme has been determined as follows: when the duration of the equipment parameter deviation from the baseline range exceeds the second duration threshold (20 minutes), or when there is a period of time within the last 24 hours during which equipment parameter adjustment processing is performed, equipment parameter adjustment processing is required.

[0156] In S31, after the equipment parameter adjustment is performed according to the change adjustment plan, the number of times each quality defect type is identified and processed during the subsequent 7-day statistical period is as follows: Type A: 42 times, Type B: 38 times, Type C: 18 times, Type D: 12 times, Type E: 45 times, Type F: 50 times, Type G: 15 times, and Type H: 48 times.

[0157] In S32, the preset threshold for the number of new additions is set to 20. Quality defect types with a number of new additions less than 20 are designated as new deviation defect types: Type C (18 times), Type D (12 times), and Type G (15 times). The number of new deviation defect types is 3.

[0158] Comparing the newly added deviation defect types with the identified risk defect types (type C, type D, type G) determined by S1: type C belongs to the identified risk defect type, type D belongs to the identified risk defect type, type G belongs to the identified risk defect type, and the number of identified risk defect types in the newly added deviation defect types is 3, all of which overlap.

[0159] In S33, the preset threshold for the number of newly added defect types is set to 4. It is determined whether 3 is greater than 4: if 3 is not greater than 4, proceed to case 2.

[0160] Among the newly added deviation defect types, there is a risk identification defect type. The risk type ratio is calculated as 3 ÷ 3 = 1.00 (all 3 newly added deviation defect types belong to the risk identification defect type). Assuming a preset risk ratio threshold of 0.60, we determine if 1.00 is greater than 0.60: if 1.00 is greater than 0.60, it is determined that active adjustment target identification processing in the production equipment is required.

[0161] In S41, acquire the equipment parameter change data of each production equipment after using the change adjustment plan, and identify the equipment parameter change period of each production equipment: Equipment 1 has 3 change periods, Equipment 2 has 2 change periods, Equipment 3 has 5 change periods, Equipment 4 has 0 change periods, Equipment 5 has 4 change periods, Equipment 6 has 0 change periods, Equipment 7 has 1 change period, Equipment 8 has 0 change periods, Equipment 9 has 3 change periods, Equipment 10 has 0 change periods, Equipment 11 has 2 change periods, and Equipment 12 has 0 change periods.

[0162] In S42, the production equipment that experiences periods of parameter variation is classified as parameter-variable production equipment: Equipment 1, Equipment 2, Equipment 3, Equipment 5, Equipment 7, Equipment 9, and Equipment 11, totaling 7 pieces of parameter-variable production equipment. The production equipment that is not classified as parameter-variable production equipment is: Equipment 4, Equipment 6, Equipment 8, Equipment 10, and Equipment 12, totaling 5 pieces of production equipment.

[0163] In S43, the percentage of production equipment with variable parameters = 7 ÷ 12 ≈ 0.583. Assuming the preset threshold for the percentage of production equipment with variable parameters is 0.50, determine if 0.583 is not less than 0.50: if 0.583 is not less than 0.50, proceed to S431.

[0164] In S431, a preset threshold for the number of variable time periods is set to 3. The number of variable time periods for each parameter of the production equipment is counted: Equipment 1 has 3 (equal to the threshold, not greater than the threshold), Equipment 2 has 2, Equipment 3 has 5 (greater than the threshold), Equipment 5 has 4 (greater than the threshold), Equipment 7 has 1, Equipment 9 has 3 (equal to the threshold), and Equipment 11 has 2. There are production equipment with more than 3 variable time periods for their parameters: Equipment 3 (5) and Equipment 5 (4), so proceed to S432.

[0165] In S432, the production equipment experiencing drastic changes is equipment 3 and equipment 5, a total of 2 units. The change matching factor is calculated as: (Number of production equipment experiencing drastic changes + Total number of production equipment with parameter changes) ÷ Total number of production equipment = (2 + 7) ÷ 12 = 9 ÷ 12 = 0.750. Assuming a preset matching factor threshold of 0.70, we determine if 0.750 is greater than 0.70: 0.750 is greater than 0.70, therefore, the production equipment whose active adjustment target is a preset proportion (assuming a preset proportion of 40%) is determined to be not a production equipment experiencing parameter changes.

[0166] There are 5 production equipment that are not among the production equipment with parameter changes (equipment 4, equipment 6, equipment 8, equipment 10, and equipment 12). 40% corresponds to 5 × 40% = 2 equipment. Therefore, 2 equipment are randomly selected as active adjustment targets. Let's assume that equipment 4 and equipment 8 are selected as active adjustment targets.

[0167] Example 2 Secondly, such as Figure 4 As shown, this invention provides an information management system for intelligent manufacturing of yoga clothing, employing the aforementioned manufacturing management method, specifically including: Adjustment control module, change adjustment module, adjustment target determination module; The adjustment and control module is responsible for determining whether production parameters need to be adjusted and managed. The change and adjustment module is responsible for determining the change and adjustment plan for the production equipment. The adjustment target determination module is responsible for determining the active adjustment target in the production equipment.

[0168] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0169] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0170] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A manufacturing management method, characterized in that, Specifically, it includes: Using the defect identification data from the image model, determine the defect identification deviation under the current production parameters. Based on the defect identification deviation under different quality defect types, determine when it is necessary to adjust and control the production parameters, and then proceed to the next step. Based on the monitoring data of the equipment parameters of the production equipment, determine the changes in the equipment parameters of different production equipment, and use the changes in the equipment parameters of different production equipment to determine the adjustment plan for the production equipment. Using the aforementioned adjustment scheme, the equipment parameters of the production equipment are adjusted when changes occur. Based on the changes in the identification data of the image model under different quality defect types, and the number of identification deviations under different quality defect types, when it is determined that active adjustment targets in the production equipment need to be identified, the active adjustment targets in the production equipment are determined based on the changes in the equipment parameters of the production equipment and the adjustment scheme.

2. The manufacturing management method as described in claim 1, characterized in that, The defect identification data includes the number of identification processes and the identification results under different quality defect types.

3. The manufacturing management method as described in claim 1, characterized in that, The defect identification deviation includes the number of identification deviations under different quality defect types.

4. The manufacturing management method as described in claim 1, characterized in that, Determine whether production parameters need to be adjusted and controlled, specifically including: Based on the defect identification deviation under different quality defect types, determine the number of identification deviations under different quality defect types; The identification deviation probability of the quality defect type is determined by using the number of identification deviations. Based on the identification deviation probability and the number of identification processes for different types of quality defects, determine whether it is necessary to adjust and control the production parameters.

5. The manufacturing management method as described in claim 4, characterized in that, The identification deviation probability of the quality defect type is determined based on the proportion of identification deviations in the number of identification processing steps for the quality defect type.

6. The manufacturing management method as described in claim 4, characterized in that, If there are no quality defect types with a probability of identification deviation greater than the preset deviation probability threshold, then it is determined that no adjustment or control management of production parameters is required.

7. The manufacturing management method as described in claim 1, characterized in that, The variation of the equipment parameters of the production equipment is determined based on the time period during which the variation rate of the equipment parameters of the production equipment meets the requirements.

8. The manufacturing management method as described in claim 1, characterized in that, The method for determining the adjustment plan for the production equipment is as follows: Based on the variation of equipment parameters of different production equipment, determine the time period in which the variation rate of the equipment parameters of the production equipment meets the requirements; The time period during which the rate of change of the equipment parameters of the production equipment meets the requirements is taken as the change matching time period of the production equipment. Based on the different time periods for the changes in production equipment, a change adjustment plan for the production equipment is determined.

9. The manufacturing management method as described in claim 8, characterized in that, If there is no matching period for the change of different production equipment, the change adjustment scheme of the production equipment is determined to be a screening adjustment scheme. That is, when the equipment parameters change, the equipment parameters need to be adjusted only when the duration of the change of the equipment parameters is not within the equipment parameter range corresponding to the baseline equipment parameters is at the first duration threshold.

10. An information management system for intelligent manufacturing of yoga clothing, employing a manufacturing management method as described in any one of claims 1-9, characterized in that, specifically... include: Adjustment control module, change adjustment module, adjustment target determination module; The adjustment and control module is responsible for determining whether production parameters need to be adjusted and managed. The change and adjustment module is responsible for determining the change and adjustment plan for the production equipment. The adjustment target determination module is responsible for determining the active adjustment target in the production equipment.

Images