Inspection mode multi-dimensional quality management system based on AI

Abstract

The invention discloses an AI-based inspection mode multi-dimensional quality management system, and belongs to the technical field of inspection supervision. The method is used for solving the technical problem that the multi-dimensional quality supervision evaluation effect and the multi-dimensional quality management effect of an inspection mode in an existing scheme are poor. Monitoring and counting inspection results of each implementation of the inspection mode, processing and classifying the inspection results of each implementation, and performing inspection failure processing of different dimensions on the abnormal inspection set to obtain a first inspection failure total value and a second inspection failure total value corresponding to different dimensions, performing data processing analysis of multi-dimensional quality management on the first inspection failure total value and the second inspection failure total value obtained by the multi-dimensional supervision processing to obtain an inspection quality value corresponding to the existing inspection mode; and performing optimization management on the inspection content of the existing inspection mode and / or performing optimization management on the existing inspection mode.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection and supervision, and particularly to a multi-dimensional quality management system for inspection methods based on AI. Background Art

[0002] Multi-dimensional quality management of inspection methods is a quality management strategy that combines multiple technologies and methods, aiming to comprehensively and systematically inspect and monitor all aspects in the production or service process to ensure that the product or service meets the predetermined quality standards. This management mode not only focuses on the quality of the final product, but also emphasizes the implementation of quality control at each stage of the production process, including raw material procurement, production processing, assembly, testing and other aspects.

[0003] When implementing the existing multi-dimensional quality management solutions for inspection methods, most of them still stay at the inspection results and management at a single level. For example, only the management of corresponding inspection defects is carried out according to the inspection results, and the existing inspection methods cannot be multi-dimensionally evaluated from different aspects, and the existing inspection methods and related inspection contents cannot be adaptively and diversely mined, analyzed and managed according to the multi-dimensional evaluation results. There are problems with poor multi-dimensional quality supervision and evaluation effects and multi-dimensional quality management effects of the inspection methods. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-dimensional quality management system for inspection methods based on AI, which is used to solve the technical problem of poor multi-dimensional quality supervision and evaluation effects and multi-dimensional quality management effects of the inspection methods in the existing solutions.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A multi-dimensional quality management system for inspection methods based on AI includes:

[0007] An inspection method implementation supervision and processing module, which is used to monitor and count the inspection results of each implementation of the inspection method, process and classify the inspection results of each implementation to obtain a normal inspection set and an abnormal inspection set, perform inspection failure processing on different dimensions of the abnormal inspection set, obtain the corresponding first total inspection failure value and the second total inspection failure value, and upload them to the cloud platform;

[0008] A multi-dimensional quality management module for inspection methods, which is used to perform data processing and analysis of multi-dimensional quality management on the first total inspection failure value and the second total inspection failure value obtained from multi-dimensional supervision and processing, obtain the inspection quality value corresponding to the existing inspection method, and optimize the management of the inspection content of the existing inspection method and / or optimize the management of the existing inspection method according to the analysis result of the inspection quality value.

[0009] Preferably, traverse and analyze the inspection results of each implementation of the inspection method. If there are no abnormal inspection items in the inspection results, mark the inspection corresponding to the inspection results as a normal inspection;

[0010] If there is at least one inspection item in the inspection results, mark the inspection corresponding to the inspection results as an abnormal inspection, and obtain the inspection components and the first abnormal reasons to which the abnormal inspection items in the abnormal inspection belong;

[0011] Sort and combine the inspection results corresponding to all normal inspections in chronological order of inspection time to obtain a normal inspection set; and sort and combine the inspection results corresponding to all abnormal inspections in chronological order of inspection time to obtain an abnormal inspection set.

[0012] Preferably, count all the inspection components and the first abnormal reasons that appear according to the abnormal inspection set, and sort and combine all the first abnormal reasons corresponding to the same inspection component to obtain an active processing sequence of inspection components;

[0013] Sort and combine all the active processing sequences of inspection components to obtain a set of active processing sequences of inspection components;

[0014] And, obtain all the equipment components with abnormalities and the corresponding second abnormal reasons, and perform data analysis on the different equipment components and their corresponding second abnormal reasons through the inspection implementation recognition model, and output the corresponding component inspection values;

[0015] The component inspection value includes a numerical value of 0, 1 or 2.

[0016] Preferably, associate the first inspection abnormality label with the equipment component to which the component inspection value with a value of 0 belongs;

[0017] Associate the second inspection abnormality label with the equipment component to which the component inspection value with a value of 1 belongs;

[0018] Associate the third inspection abnormality label with the equipment component to which the component inspection value with a value of 2 belongs;

[0019] Count all the equipment components associated with the first inspection abnormality label and mark them as the first components, and count all the equipment components associated with the second inspection abnormality label and mark them as the second components.

[0020] Preferably, calculate the total number of all the first abnormal reasons corresponding to different first components and the total number of all the second abnormal reasons matching the first abnormal reasons in sequence to obtain the corresponding first inspection failure value;

[0021] Sort and combine all the first inspection failure values, and sum up all the sorted and combined first inspection failure values to obtain the total first inspection failure value.

[0022] Preferably, the total number of all second abnormal causes corresponding to different second components is calculated to obtain the corresponding second inspection failure value;

[0023] All the second inspection failure values are sorted and combined, and the sum of all the sorted and combined second inspection failure values is obtained to get the total second inspection failure value.

[0024] Preferably, when performing multi-dimensional quality management on the inspection method, the obtained total first inspection failure value and total second inspection failure value are analyzed through an inspection quality identification model, and the corresponding inspection quality value is output;

[0025] Among them, the inspection quality value includes numerical values of 0, -1, -2 or -3.

[0026] Preferably, according to the inspection quality value of 0, it is prompted that the inspection quality of the existing inspection method is normal, and its subsequent normal implementation is maintained;

[0027] According to the inspection quality value of -1, it is prompted that the inspection quality of the existing inspection method has a first abnormality, and it is prompted to optimize the management of the inspection content corresponding to the existing inspection method;

[0028] According to the inspection quality value of -2, it is prompted that the inspection quality of the existing inspection method has a second abnormality, and it is prompted to optimize the management of the existing inspection method;

[0029] According to the inspection quality value of -3, it is prompted that the inspection quality of the existing inspection method has a third abnormality, and it is prompted to optimize the management of both the existing inspection method and the existing inspection content.

[0030] Compared with the existing solution, the beneficial effects achieved by the present invention:

[0031] By monitoring and statistically analyzing the inspection results of each implementation of the inspection method, processing and classifying the inspection results of each implementation, and performing inspection failure processing on the abnormal inspection set in different dimensions, the total first inspection failure value and total second inspection failure value corresponding to different dimensions are obtained. The present invention can not only realize the digital processing of inspection failure data of the existing inspection method from different dimensions, but also provide reliable multi-dimensional data support for the subsequent analysis of multi-dimensional quality management of the existing inspection method.

[0032] The present invention performs data processing and analysis of multi-dimensional quality management on the first total inspection failure value and the second total inspection failure value obtained through multi-dimensional supervision and processing, obtains the inspection quality value corresponding to the existing inspection method, and optimizes the inspection content of the existing inspection method and / or optimizes the existing inspection method according to the analysis result of the inspection quality value, realizing multi-dimensional supervision and processing of the existing inspection method and optimization management of multiple objects, and improving the multi-dimensional quality supervision and evaluation effect and multi-dimensional quality management effect of the inspection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 It is a block diagram of a multi-dimensional quality management system for an inspection method based on AI according to the present invention.

[0035] Figure 2 It is a flow block diagram of the operation of a multi-dimensional quality management system for an inspection method based on AI according to the present invention.

[0036] Figure 3 It is a flow block diagram for performing multi-dimensional quality management in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] As Figures 1 to 2 shown, the present invention is a multi-dimensional quality management system for an inspection method based on AI, including an inspection method implementation supervision and processing module, which is used to monitor and statistically analyze the inspection results of each implementation of the inspection method, process and classify the inspection results of each implementation, obtain a normal inspection set and an abnormal inspection set, perform inspection failure processing in different dimensions on the abnormal inspection set, obtain the corresponding first total inspection failure value and second total inspection failure value, and upload them to the cloud platform; including:

[0039] Among them, the object of the implementation of the inspection method can specifically be equipment; the specific inspection content corresponding to the inspection method can be determined according to the actual inspection equipment and actual inspection requirement data, and there is no specific limitation;

[0040] Traverse and analyze the inspection results of each implementation of the inspection method. If there are no abnormal inspection items in the inspection results, mark the inspection corresponding to the inspection results as a normal inspection;

[0041] If there is at least one inspection item in the inspection result, mark the inspection corresponding to the inspection result as an abnormal inspection, and obtain the inspection components and the first abnormal cause to which the abnormal inspection items that appear in the abnormal inspection belong; the first abnormal cause is determined by the existing verification of professional technicians in the field;

[0042] Sort and combine the inspection results corresponding to all normal inspections in the order of inspection time to obtain a normal inspection set; and sort and combine the inspection results corresponding to all abnormal inspections in the order of inspection time to obtain an abnormal inspection set;

[0043] Count all the inspection components and the first abnormal cause that appear according to the abnormal inspection set, and sort and combine all the first abnormal causes corresponding to the same inspection component to obtain an active processing sequence of inspection components;

[0044] Sort and combine all the active processing sequences of inspection components to obtain a set of active processing sequences of inspection components;

[0045] In the embodiment of the present invention, by processing and classifying all abnormal inspection data, it can provide reliable screening data support for data analysis of different aspects corresponding to all device components that have anomalies subsequently.

[0046] And, obtain all device components that have anomalies and the corresponding second abnormal cause, and perform data analysis on the different device components that appear and their corresponding second abnormal causes through an inspection implementation recognition model, and output the corresponding component inspection value BX;

[0047] Among them, the expression of the inspection implementation recognition model is In the formula, a and b are the device components and the corresponding second abnormal cause respectively; U1 is the set corresponding to all inspection components; U2 is the set of all abnormal causes of the historical inspections corresponding to the device components;

[0048] The component inspection value includes a numerical value of 0, 1 or 2;

[0049] Associate the first inspection anomaly label with the device component to which the component inspection value with a value of 0 belongs;

[0050] Associate the second inspection anomaly label with the device component to which the component inspection value with a value of 1 belongs;

[0051] Associate the third inspection anomaly label with the device component to which the component inspection value with a value of 2 belongs;

[0052] It can be understood that the first inspection anomaly label represents the corresponding equipment component and the second anomaly cause that occurred, which has been inspected historically and is the same as the first anomaly cause obtained from the inspection; the second inspection anomaly label represents the corresponding equipment component and the second anomaly cause that occurred, which has been inspected historically but is different from all the first anomaly causes obtained from the inspection; the third inspection anomaly label represents that the corresponding equipment component has not been inspected historically;

[0053] Count all the equipment components associated with the first inspection anomaly labels and mark them as the first components, and count all the equipment components associated with the second inspection anomaly labels and mark them as the second components;

[0054] In the embodiments of the present invention, by performing data processing on all the equipment components with anomalies and the corresponding second anomaly causes, the corresponding component inspection values are obtained, and different equipment components with anomalies are dynamically classified and marked according to the component inspection values, which can provide reliable anomaly classification data support for subsequent processing of inspection failure data corresponding to different dimensions in the existing inspection methods;

[0055] Through the formula Calculate the first inspection failure value XSYi corresponding to different first components; in the formula, i represents different first components, i = 1, 2, 3,..., n; n is a positive integer; represents the total number of all first components; ai is the component influence coefficient corresponding to different first components, and the specific value can be determined by professional technical personnel in the field according to inspection experience, or can be determined according to the total number of historical inspection anomalies corresponding to different first components; Ni is the total number of all second anomaly causes that match the first anomaly cause corresponding to different first components; N0i is the total number of all first anomaly causes corresponding to different first components;

[0056] Sort and combine all the first inspection failure values, and sum up all the sorted and combined first inspection failure values to obtain the total first inspection failure value XSY0;

[0057] And, through the formula XSEj = bj × e N1j Calculate the second inspection failure value XSEj corresponding to different second components; in the formula, j represents different second components, j = 1, 2, 3,..., m; m is a positive integer; represents the total number of all second components; bj is the component influence coefficient corresponding to different second components, and the specific value can be determined by professional technical personnel in the field according to inspection experience, or can be determined according to the total number of historical inspection anomalies corresponding to different first components; N1j is the total number of all second anomaly causes corresponding to different second components; e is a constant;

[0058] Sort and combine all the second patrol failure values, and sum up all the sorted and combined second patrol failure values to obtain the total second patrol failure value XSE0;

[0059] It should be noted that the total first patrol failure value and the total second patrol failure value are used to digitally represent the corresponding patrol failure impacts from different types of patrol failures;

[0060] In the embodiments of the present invention, by monitoring and statistically analyzing the patrol results of each implementation of the patrol method, processing and classifying the patrol results of each implementation, and performing patrol failure processing on the abnormal patrol set in different dimensions, the total first patrol failure value and the total second patrol failure value corresponding to different dimensions are obtained. It can not only realize the digital processing of patrol failure data for the existing patrol method from different dimensions, but also provide reliable multi-dimensional data support for the subsequent analysis of multi-dimensional quality management of the existing patrol method.

[0061] The multi-dimensional quality management module for the patrol method is used to perform data processing and analysis of multi-dimensional quality management on the total first patrol failure value and the total second patrol failure value obtained from multi-dimensional supervision and processing, obtain the patrol quality value corresponding to the existing patrol method, and optimize the management of the patrol content of the existing patrol method and / or optimize the management of the existing patrol method according to the analysis result of the patrol quality value; including:

[0062] As Figure 3 shown, when performing multi-dimensional quality management on the patrol method, the total first patrol failure value and the total second patrol failure value obtained through processing are analyzed through the patrol quality recognition model, and the corresponding patrol quality value XZ is output;

[0063] Among them, the expression of the patrol quality recognition model is In the formula, A, B, and C are the first patrol failure limit value, the second patrol failure limit value, and the third patrol failure limit value respectively. The specific numerical values can be determined according to the actual patrol requirement data, or can be determined according to the test data in the early stage of the implementation of the existing patrol method. The specific numerical values are not limited; XSS0 is the third patrol failure value;

[0064] The third patrol failure value is calculated through the formula ; in the formula, N2 is the total number of occurrences corresponding to the third patrol abnormal label; NZ is the total number of all patrol components for which the existing patrol method conducts patrols;

[0065] The patrol quality value includes numerical values of 0, -1, -2, or -3;

[0066] It should be noted that by integrating and processing the digital data of the inspection failure handling in different dimensions in the early stage, the corresponding inspection quality value is obtained. During the calculation of the inspection quality value, due to the processing and calculation of the inspection method and inspection content, subsequent analysis and utilization in the corresponding aspects can be realized, and multi-dimensional quality management of the inspection method and inspection content can be achieved.

[0067] According to the inspection quality value of 0, it is prompted that the inspection quality of the existing inspection method is normal, and its subsequent normal implementation is maintained.

[0068] According to the inspection quality value of -1, it is prompted that the inspection quality of the existing inspection method is the first anomaly, and it is prompted to optimize the management of the inspection content corresponding to the existing inspection method.

[0069] According to the inspection quality value of -2, it is prompted that the inspection quality of the existing inspection method is the second anomaly, and it is prompted to optimize the management of the existing inspection method.

[0070] According to the inspection quality value of -3, it is prompted that the inspection quality of the existing inspection method is the third anomaly, and it is prompted to optimize the management of both the existing inspection method and the existing inspection content.

[0071] Among them, to optimize the management of the inspection content, specifically, it can be to add, delete, or modify on the basis of the existing inspection content; to optimize the management of the existing inspection method, specifically, it can be to replace the existing inspection method, or to synchronously perform the existing inspection method in combination with other inspection methods, which can be customized according to the actual application requirements of the actual application scenario.

[0072] It should be noted that by performing data analysis on the inspection quality value obtained through integration and processing, optimization management at the inspection content level and inspection method level can be realized, and diversified processing, analysis, and utilization of the inspection quality value are achieved.

[0073] In the embodiments of the present invention, through multi-dimensional quality management data processing and analysis of the first total inspection failure value and the second total inspection failure value obtained through multi-dimensional supervision processing, the inspection quality value corresponding to the existing inspection method is obtained. According to the analysis result of the inspection quality value, the inspection content of the existing inspection method is optimized and managed and / or the existing inspection method is optimized and managed, realizing multi-dimensional supervision processing of the existing inspection method and optimization management of multiple objects, and improving the multi-dimensional quality supervision and evaluation effect and multi-dimensional quality management effect of the inspection method.

[0074] In several embodiments provided by the present invention, it should be understood that the disclosed system can be implemented in other ways. For example, the above-described invention embodiments are merely illustrative. For example, the division of modules is only a logical function division, and there can be other division methods in actual implementation.

[0075] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0076] In addition, in each embodiment of the present invention, the functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of a combination of hardware and software functional modules.

[0077] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the basic features of the present invention, the present invention can be implemented in other specific forms.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An AI-based multi-dimensional quality management system for inspection methods, characterized in that, Including: An inspection method implementation supervision and processing module, which is used to monitor and count the inspection results of each implementation of the inspection method, process and classify the inspection results of each implementation to obtain a normal inspection set and an abnormal inspection set, perform inspection failure processing on different dimensions of the abnormal inspection set, obtain the corresponding first inspection failure total value and second inspection failure total value, and upload them to the cloud platform; An inspection method multi-dimensional quality management module, which is used to perform data processing and analysis of multi-dimensional quality management on the first inspection failure total value and the second inspection failure total value obtained from multi-dimensional supervision and processing, obtain the inspection quality value corresponding to the existing inspection method, and optimize the management of the inspection content of the existing inspection method and / or optimize the management of the existing inspection method according to the analysis result of the inspection quality value.

2. The multi-dimensional quality management system for the AI-based inspection method according to claim 1, characterized in that, Traverse and analyze the inspection results of each implementation of the inspection method. If there are no abnormal inspection items in the inspection results, mark the inspection corresponding to the inspection results as a normal inspection; If there is at least one inspection item in the inspection results, mark the inspection corresponding to the inspection results as an abnormal inspection, and obtain the inspection components and the first abnormal reason to which the abnormal inspection items in the abnormal inspection belong; Sort and combine the inspection results corresponding to all normal inspections in the order of inspection time to obtain a normal inspection set; And sort and combine the inspection results corresponding to all abnormal inspections in the order of inspection time to obtain an abnormal inspection set.

3. The multi-dimensional quality management system for the AI-based inspection method according to claim 2, characterized in that, Count all the inspection components and the first abnormal reasons that appear according to the abnormal inspection set, and sort and combine all the first abnormal reasons corresponding to the same inspection component to obtain an inspection component active processing sequence; Sort and combine all the inspection component active processing sequences to obtain an inspection component active processing sequence set; And obtain all the device components with abnormalities and the corresponding second abnormal reasons, perform data analysis on the different device components and their corresponding second abnormal reasons through an inspection implementation identification model, and output the corresponding component inspection value; The component inspection value includes a numerical value of 0, 1, or 2.

4. The multi-dimensional quality management system for the AI-based inspection method according to claim 3, wherein, Associate the first inspection abnormality label with the device component to which the component inspection value with a numerical value of 0 belongs; Associate the second inspection abnormality label with the device component to which the component inspection value with a numerical value of 1 belongs; Associate the third inspection abnormality label with the device component to which the component inspection value with a numerical value of 2 belongs; Count all the device components corresponding to the associated first inspection abnormality labels and mark them as the first components, and count all the device components corresponding to the associated second inspection abnormality labels and mark them as the second components.

5. The multi-dimensional quality management system for AI-based inspection methods according to claim 4, wherein, Calculate the corresponding first inspection failure value in sequence for the total number of all the first abnormal reasons corresponding to different first components and the total number of all the second abnormal reasons matching the first abnormal reasons; Sort and combine all the first inspection failure values, and sum up all the sorted and combined first inspection failure values to obtain the first inspection failure total value.

6. The multi-dimensional quality management system for AI-based inspection methods according to claim 5, characterized in that, Calculate the corresponding second inspection failure value for the total number of all the second abnormal reasons corresponding to different second components; Sort and combine all the second inspection failure values, and sum up all the sorted and combined second inspection failure values to obtain the second inspection failure total value.

7. The multi-dimensional quality management system for AI-based inspection methods according to claim 6, characterized in that, When performing multi-dimensional quality management on the inspection method, the first total inspection failure value and the second total inspection failure value obtained are analyzed through the inspection quality identification model, and the corresponding inspection quality value is output; Among them, the inspection quality value includes numerical values of 0, -1, -2, or -3.

8. The multi-dimensional quality management system for AI-based inspection methods according to claim 7, characterized in that, According to the inspection quality value of 0, it is prompted that the inspection quality of the existing inspection method is normal, and its subsequent normal implementation is maintained; According to the inspection quality value of -1, it is prompted that the inspection quality of the existing inspection method is the first anomaly, and it is prompted to optimize the management of the inspection content corresponding to the existing inspection method; According to the inspection quality value of -2, it is prompted that the inspection quality of the existing inspection method is the second anomaly, and it is prompted to optimize the management of the existing inspection method; According to the inspection quality value of -3, it is prompted that the inspection quality of the existing inspection method is the third anomaly, and it is prompted to optimize the management of both the existing inspection method and the existing inspection content.

Images