Automobile production quantitative closed-loop management and control method and system based on active feedback mechanism

Abstract

The invention is suitable for the technical field of industrial automation and control, and provides an automobile production quantitative closed-loop management and control method and system based on an active feedback mechanism. The system comprises a plan management module, an index target planning module, an execution auxiliary module, an active feedback module, a process supervision module, an inspection correction module, a problem solving module and a knowledge base module. The method can reduce the plan execution deviation rate, improve the problem response speed, reduce the information processing load of production management and control activities, improve the accuracy, timeliness and scientificity of production management and control, and is suitable for refined quality management and control of complex scenes.

Description

Technical Field

[0001] The present invention relates to the field of industrial automation and control technology, and in particular to a quantitative closed-loop control method and system for automobile production based on an active feedback mechanism. Background Art

[0002] The PDCA cycle, or "Plan-Do-Check-Act," is a continuous improvement management approach originally proposed by quality management expert Walter Armand Shewhart. It was further developed and popularized by Japanese quality management expert Deming in the 1950s, earning it the nickname the Deming Cycle. It serves as the ideological foundation and methodological basis for total quality management. This cyclical approach is widely used across various organizations and industries to improve and optimize processes, products, and services. The PDCA cycle is an ongoing process, with each stage interconnected and dependent on the next. Through an iterative approach, organizations can continuously identify problems, develop solutions, implement improvements, and evaluate their effectiveness, thereby achieving continuous improvement and optimization. This approach emphasizes data-driven, fact-based decision-making, as well as teamwork and cross-departmental collaboration, making it a crucial tool in modern management practices.

[0003] The PDCA cycle consists of the following four stages: Plan: During this phase, improvement goals and objectives are determined. Problems or opportunities are first identified and analyzed, and then plans are developed to address the problems or exploit the opportunities. The goal is to ensure that the plans are clear, feasible, and aligned with the organization's strategic goals.

[0004] Do During this phase, implementation is carried out according to the plan. This may involve developing and implementing new processes, methods, or activities, or modifying existing processes. During implementation, data and information may be collected for analysis during the subsequent review and evaluation phases.

[0005] Check: During this phase, we evaluate the results and effectiveness of implementation and compare them with the goals set in the plan. By comparing actual results with expected results, we determine whether the goals were achieved. This phase emphasizes data analysis and evaluation to identify problems, bottlenecks, or areas for improvement.

[0006] Act Based on the results of the review phase, appropriate actions are taken. If the goals have been achieved, the improved results can be standardized and fixed; if not, the plan needs to be adjusted and re-executed. The goal of this phase is continuous improvement and optimization to ensure continuous improvement in the organization's performance and efficiency. Summary of the Invention

[0007] The present invention provides a quantitative closed-loop control method and system for automobile production based on an active feedback mechanism. This method breaks down qualitative plans into quantitative control indicators and targets, improving the accuracy and controllability of the plans. The closed-loop feedback mechanism is used to ensure the implementation of the plans. Combined with the support of information systems, the information processing load during the control process is reduced, thereby improving the quantification, visibility, and execution efficiency of control activities.

[0008] A quantitative closed-loop control system for automobile production based on an active feedback mechanism, the system comprising: The plan management module is used to formulate and break down plans, quantify tasks using a matrix-based quantification method, establish a work list P, and continuously optimize it. The indicator target planning module is used to establish an evaluation indicator system M with weights based on the work list and build a multi-dimensional target matrix with embedded time schedule , quantify work content through matrix indicator goals; Execution assistance module, used to dynamically push work specifications and operating standards based on the knowledge base; Active feedback module, used to collect operation indicator data and abnormal problem information in real time and report the data; A process monitoring module to verify process compliance through inspections and trigger corrective actions; The inspection and correction module is used to calculate the comprehensive deviation value based on the deviation analysis matrix C and trigger the correction strategy or plan revision according to the threshold.

[0009] Preferably, the work item list in the plan management module , where k is the number of items, is the i-th item.

[0010] Preferably, the indicator target planning module includes: The evaluation index system construction unit is established by using vectorized weight distribution method , where n is the number of indicators, is the i-th indicator; Target benchmark generation unit, building a multi-dimensional target matrix based on the time dimension , where t is the number of all time points considered, n is the total number of all indicators, represents the target value of the jth indicator at the i-th time node; Comprehensive evaluation calculation unit, execute the formula: , where n is the number of indicators, is the weight of the jth indicator at the i-th time node, Indicates the target value of the jth indicator at the i-th time node.

[0011] Preferably, the inspection and correction module includes: Deviation analysis unit, the deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target difference of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ; in, is the weight of the jth indicator at the i-th time node; Threshold determination unit, control threshold ratio is , using three-layer judgment logic: the calculation formula is as follows: ; in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ; when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted.

[0012] Preferably, the active feedback mechanism in the active feedback module is implemented by setting an automatic triggering reporting protocol for abnormal data, and automatically generating a feedback report containing a deviation data packet and associated responsible parties when it is detected that the indicator fluctuation exceeds a preset threshold parameter.

[0013] Preferably, the system further comprises: The problem-solving module is used to locate the root cause of the problem through responsibility tracing and to track the solution process in a closed loop; The knowledge base module is used to store and iteratively update the standard knowledge graph including the solution library and specification library.

[0014] Preferably, the optimization mechanism of the knowledge base module includes: Build a case-solution mapping model and use similarity matching algorithms to associate historical issues with solutions; Set knowledge iteration rules. When the same problem occurs more than 3 times N times, the standard specification revision process is automatically triggered.

[0015] Another object of the present invention is to provide a quantitative closed-loop control method for automobile production based on an active feedback mechanism, the method comprising: Formulate and break down plans, quantify tasks using matrix-based quantitative methods, establish a task list P, and continuously optimize it; Establish an evaluation index system M with weights based on the work list, and build a multi-dimensional target matrix with embedded time schedule , quantify work content through matrix indicator goals; Dynamically push work specifications and operating standards based on the knowledge base; Collect operational indicator data and abnormal problem information in real time and report the data; Verify process compliance through inspections and trigger corrective actions; The comprehensive deviation value is calculated based on the deviation analysis matrix C, and the correction strategy or plan revision is triggered according to the threshold.

[0016] Preferably, the evaluation index system M including weights is established based on the work list, and a multi-dimensional target matrix with embedded time schedule is constructed. , quantifying the work content through matrix indicators and targets includes the following steps: Step 2.1: Determine the work priorities based on the work items; Step 2.2: Set up a work performance evaluation index system. The evaluation index system M is a vector containing multiple indicators, where each indicator has a different weight. The evaluation index system M is represented by a vector: ; Where n is the number of indicators, is the i-th indicator; Step 2.3: Generate a work item plan based on the work item list and evaluation indicator system; Step 2.4: Determine the multidimensional target matrix of expected results based on the evaluation indicator system and time schedule. The multidimensional target matrix O is composed of the target values of indicators based on time points and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target value of the jth indicator at the i-th time node; Comprehensive evaluation objectives The calculation formula is as follows: ; Where n is the number of indicators, is the weight of the jth indicator at the i-th time node; Step 2.5: Generate a work plan based on indicators and goals.

[0017] Preferably, the step of calculating the comprehensive deviation value based on the deviation analysis matrix C and triggering the deviation correction strategy or plan revision according to the threshold value comprises the following steps: Step 6.1: Check the deviation between the work results and the expected goals; Step 6.2: Generate a deviation analysis report, calculate the deviation analysis matrix, and evaluate whether the deviation exceeds the preset threshold. The deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target difference of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ; in, is the weight of the jth indicator at the i-th time node; Step 6.3: If there is a deviation but it does not exceed the control threshold, a correction activity is triggered and the corresponding treatment plan is executed according to the preset solution library. If there is a deviation but it exceeds the control threshold, a plan modification activity is triggered to adjust the work content plan. If there is no deviation, the process ends. This rule is expressed as the following formula: The control threshold ratio is , the calculation formula is as follows: ; in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ; when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted; Step 6.4: Develop a correction plan based on the solution library preset in the knowledge base; Step 6.5: Issue correction instructions; Step 6.6: Receive correction instructions and carry out correction activities.

[0018] Beneficial effects achieved by the present invention: Quantitative management: Based on a more scientific management approach, this system transforms corporate plans from qualitative descriptions into quantitative plans, condenses core requirements and work expectations into indicators and goals, and achieves refined management, thereby improving the accuracy, effectiveness and timeliness of management and control activities.

[0019] Closed-loop management: Each part of the system forms a continuous closed loop, and provides timely and effective information feedback and makes corresponding adjustments based on the dynamic changes in objective reality, promptly resolving anomalies and problems in the operation process, and accumulating experience for continuous optimization, thereby continuously improving in the cycle accumulation and promoting the self-renewal and development of the system.

[0020] Active Feedback: Compared to traditional production control systems, this system introduces an active feedback mechanism, enabling workers at every level to proactively report work results and anomalies to direct management. This innovation eliminates the traditional step of proactive inquiry by middle and upper-level managers, reduces the amount of information processed, and thus effectively reduces the system's information processing load, improving manager efficiency and decision-making speed, allowing managers to focus more on more important tasks.

[0021] Knowledge Optimization: This system uses a knowledge base to drive operational process optimization. During operation, the system automatically collects information, records issues, and distills this information into knowledge, which is stored in the knowledge base. Based on the data in the knowledge base, the system iteratively optimizes production processes and operating standards, effectively improving work efficiency and preventing the recurrence of similar issues. This system enables companies to better leverage existing experience, continuously improve production processes, and enhance overall operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of a quantitative closed-loop control system for automobile production based on an active feedback mechanism provided by an embodiment of the present invention.

[0023] Figure 2 It is a flow chart of a quantitative closed-loop control method for automobile production based on an active feedback mechanism provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0026] See also Figure 1 , an embodiment of the present invention provides a quantitative closed-loop control system for automobile production based on an active feedback mechanism, the system comprising: The plan management module is used to formulate and break down plans, quantify tasks using a matrix-based quantification method, establish a work list P, and continuously optimize it. The indicator target planning module is used to establish an evaluation indicator system M with weights based on the work list and build a multi-dimensional target matrix with embedded time schedule , quantify work content through matrix indicator goals; Execution assistance module, used to dynamically push work specifications and operating standards based on the knowledge base; Active feedback module, used to collect operation indicator data and abnormal problem information in real time and report the data; A process monitoring module to verify process compliance through inspections and trigger corrective actions; The inspection and correction module is used to calculate the comprehensive deviation value based on the deviation analysis matrix C and trigger the correction strategy or plan revision according to the threshold; The problem-solving module is used to locate the root cause of the problem through responsibility tracing and to track the solution process in a closed loop; The knowledge base module is used to store and iteratively update the standard knowledge graph including the solution library and specification library.

[0027] In this embodiment, the work item list in the plan management module , where k is the number of items, is the i-th item.

[0028] In this embodiment, the indicator target planning module includes: The evaluation index system construction unit is established by using vectorized weight distribution method ; Target benchmark generation unit, building a multi-dimensional target matrix based on the time dimension , where t is the number of all time points considered, n is the total number of all indicators, represents the target value of the jth indicator at the i-th time node; Comprehensive evaluation calculation unit, execute the formula: , where n is the number of indicators, is the weight of the jth indicator at the i-th time node, Indicates the target value of the jth indicator at the i-th time node.

[0029] In this embodiment, the inspection and correction module includes: Deviation analysis unit, the deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target difference of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ; in, is the weight of the jth indicator at the i-th time node; Threshold determination unit, control threshold ratio is , using three-layer judgment logic: the calculation formula is as follows: ; in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ; when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted.

[0030] In this embodiment, the active feedback mechanism in the active feedback module is implemented by setting an automatic triggering reporting protocol for abnormal data, and automatically generating a feedback report containing a deviation data packet and associated responsible parties when it is detected that the indicator fluctuation exceeds a preset threshold parameter.

[0031] In this embodiment, the optimization mechanism of the knowledge base module includes: Build a case-solution mapping model and use similarity matching algorithms to associate historical issues with solutions; Set knowledge iteration rules. When the same problem occurs more than 3 times N times, the standard specification revision process is automatically triggered.

[0032] Another object of the present invention is to provide a quantitative closed-loop control method for automobile production based on an active feedback mechanism, the method comprising: Step 1: Plan formulation and breakdown, quantify tasks using a matrix-based quantification method, establish a task list P, and continuously optimize it; Step 2: Establish an evaluation index system M with weights based on the work list, and construct a multi-dimensional target matrix with embedded time schedules , quantify work content through matrix indicator goals; Step 3: Dynamically push work specifications and operating standards based on the knowledge base; Step 4: Collect operational indicator data and abnormal problem information in real time and report the data; Step 5: Verify process compliance through inspections and trigger corrective actions; Step 6: Calculate the comprehensive deviation value based on the deviation analysis matrix C and trigger the correction strategy or plan revision according to the threshold; Step 7: Identify the root cause of the problem through responsibility tracing and track the resolution process in a closed-loop manner; Step 8: Store and iteratively update the standard knowledge graph containing the solution library and specification library.

[0033] In this embodiment, the formulation and breakdown of the plan, vectorizing the work items through a matrix quantization method, establishing a work item list P and continuously optimizing it include the following steps: Step 1.1: Determine the work content; Step 1.2: Decompose the work tasks based on the automobile production process, including vertical decomposition between levels and horizontal decomposition between different items, and formulate a work list. The work list P is represented by a vector: ; Where k is the number of items, is the i-th item.

[0034] Step 1.3: Update your to-do list.

[0035] In this embodiment, the evaluation index system M including weights is established based on the work list, and a multi-dimensional target matrix with embedded time schedule is constructed. , quantifying the work content through matrix indicators and targets includes the following steps: Step 2.1: Determine the work priorities based on the work items; Step 2.2: Set up a work performance evaluation index system. The evaluation index system M is a vector containing multiple indicators, where each indicator has a different weight. The evaluation index system M is represented by a vector: ; Where n is the number of indicators, is the i-th indicator; Step 2.3: Generate a work item plan based on the work item list and evaluation indicator system; Step 2.4: Determine the multidimensional target matrix of expected results based on the evaluation indicator system and time schedule. The multidimensional target matrix O is composed of the target values of indicators based on time points and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target value of the jth indicator at the i-th time node; Comprehensive evaluation objectives The calculation formula is as follows: ; Where n is the number of indicators, is the weight of the jth indicator at the i-th time node; Step 2.5: Generate a work plan based on indicators and goals.

[0036] In this embodiment, dynamically pushing work specifications and operating standards based on the knowledge base includes the following steps: Step 3.1: Issue work instructions.

[0037] Step 3.2: The task instruction information is sent to the corresponding task responsible person.

[0038] Step 3.3: Retrieve the work specifications and operating standards related to the task from the knowledge base and send them to the person responsible for the task to provide assistance for work execution.

[0039] In this embodiment, real-time collection of job indicator data and abnormal problem information and reporting of the data includes the following steps: Step 4.1: Report operational metrics and work results; Step 4.2: Feedback on exceptions and problems encountered during job execution; Step 4.4: Provide guidance or solutions for abnormal issues.

[0040] In this embodiment, checking and verifying process compliance and triggering rectification tasks include the following steps: Step 5.1: Check whether the execution layer follows the work specifications and operating standards. Specific methods include spot inspection, on-site inspection, video inspection, etc. Step 5.2: If there is a problem, issue a corrective action.

[0041] In this embodiment, the calculation of the comprehensive deviation value based on the deviation analysis matrix C and triggering the deviation correction strategy or plan revision according to the threshold value include the following steps: Step 6.1: Check the deviation between the work results and the expected goals; Step 6.2: Generate a deviation analysis report, calculate the deviation analysis matrix, and evaluate whether the deviation exceeds the preset threshold. The deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ; Where t is the number of time points, n is the number of indicators, is the target difference of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ; in, is the weight of the jth indicator at the i-th time node; Step 6.3: If there is a deviation but it does not exceed the control threshold, a correction activity is triggered and the corresponding treatment plan is executed according to the preset solution library. If there is a deviation but it exceeds the control threshold, a plan modification activity is triggered to adjust the work content plan. If there is no deviation, the process ends. This rule is expressed as the following formula: The control threshold ratio is , the calculation formula is as follows: ; in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ; when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted; Step 6.4: Develop a correction plan based on the solution library preset in the knowledge base; Step 6.5: Issue correction instructions; Step 6.6: Receive correction instructions and carry out correction activities.

[0042] In this embodiment, locating the root cause of the problem through responsibility tracing and closed-loop tracking of the resolution process includes the following steps: Step 7.1: Identify and analyze the causes of abnormal problems; Step 7.2: Identify the responsible department and relevant persons responsible for the problem; Step 7.3: Assign the problem-solving task to the appropriate personnel; Step 7.4: Carry out problem solving; Step 7.5: Review the progress results of the problem-solving work to ensure that the problem has been effectively resolved and meets the expected quality standards.

[0043] In this embodiment, storing and iteratively updating the standard knowledge graph including the solution library and the specification library includes the following steps: Step 8.1: Create and manage work specifications and operating standards; Step 8.2: Update and revise work specifications and operating standards The present invention provides an embodiment in which a control process based on a multi-layer closed-loop production control method includes the following steps: Step 1: Determine the work content and make a work list.

[0044] Step 2: Set up a work performance evaluation index system, develop a work schedule, and include the index system in the knowledge base.

[0045] Step 3: Determine the expected work goals based on the indicator system and time points, develop a work plan based on the indicators and goals, and include the work plan in the knowledge base.

[0046] Step 4: Deliver work instructions, provide workflow and operating standards in the knowledge base, and carry out work according to the process standards.

[0047] Step 5: Actively report work results and provide feedback on work anomalies and problems.

[0048] Step 6: Monitor the implementation of work specifications and standard operations.

[0049] Step 7: Check for deviations between the actual work results and the work plan, and generate a work plan deviation analysis report. If a deviation exists but does not exceed the threshold, proceed to Step 8 and implement corrective actions. If a deviation exists and exceeds the threshold, proceed to Step 1 and revise the work content based on the work progress and knowledge base. If no deviation exists, the process ends.

[0050] Step 8: Based on the work plan table, work plan deviation analysis report and knowledge base preset solution library, formulate a correction strategy and issue correction instructions.

[0051] Step 9: Carry out corrective activities, implement the corrective treatment plan, and initiate step 5, which is to report the work results and problems of the corrective activities.

[0052] Step 10: Analyze abnormal problems and carry out problem solving.

[0053] Step 11: Record the correction process and solutions, summarize the work results and experience, update the knowledge base, and improve the workflow specifications and operating standards based on the knowledge base to continuously improve the workflow.

[0054] like Figure 2 As shown, the present invention provides an embodiment of a method for quantitative closed-loop control of automobile production based on an active feedback mechanism, comprising the following steps: The production decision-making level determines the company-level work plan, the production management level determines the department-level work plan, and the production control level determines the workshop-level work plan. Each level formulates a list of work items.

[0055] The production decision-making level develops a work performance evaluation index system, the production management level determines department-level performance indicators, and the production control level determines workshop-level performance indicators. Each level develops a work schedule and incorporates the index system into the knowledge base.

[0056] Each level determines expected work goals based on a set of indicators and timelines. Specifically, the production decision-making level determines company-level goals, the production management level determines department-level goals, and the production control level determines shop-level goals. Each level generates a work plan based on these indicators and goals and stores it in the knowledge base.

[0057] The production control layer issues work instructions, and the production operation layer executes them. Work specifications and operating standards are retrieved from the knowledge base and sent to task owners in the production operation layer, who then carry out their work according to the process standards.

[0058] The task responsible persons in the production operation layer proactively report the work results to the task responsible persons in the production control layer and provide feedback on work anomalies and problems.

[0059] The task responsible persons in the production control layer supervise the actual implementation of work specifications and standard operations.

[0060] The task responsible person in the production control layer checks whether there are any deviations between the actual work results and the work plan, and generates a work plan deviation analysis report. If there is a deviation and it exceeds the threshold, proceed to step 8; if there is a deviation but it does not exceed the threshold, proceed to step 10; if there is no deviation, the process ends.

[0061] The production operation layer reports the work plan deviation analysis report to the production control layer. The task responsible person in the production control layer conducts scheduling control deviation. The production decision layer checks whether there is a deviation. If there is no deviation, the process ends; if there is a deviation, execute step 9.

[0062] The production decision-maker adjusts the work plan to control the deviation and executes step 1.

[0063] The task responsible person in the production control layer formulates a correction strategy based on the work plan table, work plan deviation analysis report and knowledge base preset solution library, and issues correction instructions to the production operation layer.

[0064] The task responsible persons in the production operation layer carry out corrective activities, implement the corrective treatment plan, and initiate step 5, which is to report the work results and problems of the corrective activities.

[0065] The responsible person in the production control layer analyzes abnormal problems, conducts problem solving, identifies and analyzes the causes of abnormal problems, and determines the responsible department and relevant persons responsible for the problems. They carry out problem solving work to ensure that the problems are effectively resolved and meet the expected quality standards.

[0066] Record the correction process and solutions, summarize work results and experience, and update the knowledge base. Based on the knowledge base, improve workflow specifications and operating standards to continuously improve the workflow.

[0067] In order to facilitate those skilled in the art to better understand the technical solutions of the present invention, specific embodiments of the present invention are given as follows: Embodiment 1: This embodiment of the present invention relates to the field of automobile production, specifically a closed-loop optimization method for automobile parts production quality control, comprising the following steps: Step 1: The production and operations management team determines the workshop's work content and develops a production checklist, identifying the core quality control task as "stamping part dimensional accuracy control." This checklist consists of: [Incoming material inspection, initial in-process inspection, process inspection, and finished product inspection]. The knowledge base is initialized, building a quality knowledge graph based on TS16949 clauses and process standard documents. A solution library is also developed, including a plan for handling excessive scrap rates and a user feedback process.

[0068] Step 2: The production and operations control team develops a workshop production schedule based on the work items and determines workshop-level performance indicators. Key evaluation indicators include external scrap rate (≤0.8% for the stamping workshop and ≤0.5% for the welding workshop), customer quality complaint response time (≤2 hours), 2TP audit and rectification completion rate (≥95%), and inspection accuracy (≥99.2% for the visual inspection system). Auxiliary evaluation indicators include the "Three Guarantees" claim rate (monthly year-on-year increase ≤5%) and quality cost control rate (≤3.8% of manufacturing cost). A workshop-level performance indicator system is constructed based on these evaluation indicators and incorporated into the knowledge base.

[0069] Step 3: The production and operations control team determines workshop production targets based on the indicator system and timelines, such as zero defective parts in post-processing, a 0.1% processing rate, a 0.2% LC defect rate, a 93% availability rate, 100% production plan achievement, and a 100% equipment preventive implementation rate. A workshop work plan based on these indicators and goals is developed, and production instructions are issued to the production operations team. The work plan is then stored in the knowledge base.

[0070] Step 4: The production operations layer receives the work order, and the system pushes the inspection task to the quality inspector's handheld terminal. The system automatically retrieves the GD&T Inspection Specifications and SPC Operation Manual from the knowledge base to guide the inspection. The operator reviews the workflow and operating standards in the knowledge base and performs the work according to the process standards.

[0071] Step 5: The production operations team proactively reports work results, uploads inspection data in real time, and summarizes any anomalies and issues to the production operations control team, completing a work observation form. The system collects real-time inspection data from welding stations. If the scrap rate increases from 0.5% to 0.63% (the threshold is 0.6%), the feedback module automatically generates an anomaly report, triggering an early warning signal that is sent to the quality manager's mobile device.

[0072] Step 6: The production and operation control layer checks the work plan matrix in the knowledge base, supervises the implementation of work specifications and standard operations, and checks system warning information.

[0073] Step 7: The inspection and correction module calculates the deviation matrix and, based on the preset rules, finds that the error rate exceeds the threshold of 0.1%. The system determines that manual intervention is required, generates a deviation analysis report, and pushes candidate solutions: Option A: Electrode cap replacement (historical success rate 82%).

[0074] Solution B: Welding current adjustment + pressurization time optimization (historical success rate 91%).

[0075] Step 8: Based on the workshop production plan matrix, deviation analysis report and knowledge base preset solution library, the production operation layer selects correction solution A and issues instructions, and executes the auxiliary module to push the "Electrode Cap Replacement Operation Guidelines".

[0076] Step 9: The production operations layer receives the instruction to carry out corrective actions, implements the corrective action plan, and initiates step 5. This means reporting the corrective action results and issues to the production operations control layer.

[0077] Step 10: The production and operation control layer analyzes the abnormal problem and uses the SPC chart to locate the abnormality occurring in the right side welding robot of station 7. The knowledge base automatically matches historical cases and finds that a similar problem in 2023 was caused by electrode cap wear. The system automatically assigns the person in charge to Engineer Wang from the Equipment Department to solve the problem.

[0078] Step 11: After the repair is completed, the system verifies that the external scrap rate has returned to 0.48%, records the correction process and solution, automatically generates a case report and stores it in the knowledge base, updates the electrode cap replacement cycle standard, incorporates the work results and experience into the knowledge base, and improves the workflow specifications and operating standards based on the knowledge base.

[0079] Example 2: This example relates to the field of automobile production, specifically a production quality control closed-loop optimization system applied to automobile painting processes, including the following steps: Step 1: The production and operations management team determines the workshop's work content and develops a production checklist, identifying the core quality control task as "stamping part dimensional accuracy control." The checklist includes: [substrate pretreatment inspection, initial spray parameter inspection, film thickness inspection, and complete finished product appearance inspection]. The knowledge base is initialized, building a quality knowledge graph based on the paint defect map, including a defect case library, rework standards, and a color difference adjustment case library.

[0080] Step 2: Develop a workshop production schedule based on the work items. Core control indicators are identified as customer quality complaint response time (≤1 hour) and first-pass rework pass rate (≥90%). Monitoring indicators are paint color difference ΔE value (≤0.8) and clearcoat film thickness (90-110μm). A workshop-level performance indicator system is constructed based on these evaluation indicators and incorporated into the knowledge base.

[0081] Step 3: Based on the indicator system and timelines, establish production targets for the workshop: customer quality complaint response time of no more than one hour, first-pass rework pass rate exceeding 90%, paint color difference ΔE value controlled within 0.8, and clearcoat film thickness within the range of 90-110μm. Based on these indicators and timelines, develop a detailed work schedule and issue production instructions to the production operations team. Simultaneously, incorporate the work schedule and related indicator system into the knowledge base.

[0082] Step 4: The production operation layer receives work instructions, including specific production tasks and quality control requirements. The production operation layer reviews the workflow and operating standards in the knowledge base to ensure that all operations comply with established specifications and standards, carry out work according to the work schedule and operating standards, and strictly control the paint surface quality and process parameters. The information sent by the system to the operation station includes: a) Film thickness detection point matrix (9-point detection method, key areas encrypted to 5×5 grid).

[0083] b) AR glasses project the standard color palette Lab* values (L=32.5, a=0.8, b=15.2) for real-time comparison.

[0084] c) Spraying robot adaptive compensation parameter package (including the corrected TCP coordinate system offset).

[0085] Step 5: The production operations team regularly reports work results and production progress to the production operations control team, including the amount of work completed and quality data (such as paint color difference ΔE values and clearcoat film thickness). Any anomalies or problems discovered during the operation are promptly summarized and reported to the production operations control team. In one case, the production operations team reported a color difference issue on the hood paint, uploaded the inspection data, and summarized the anomalies and problems to the production operations control team, who then filled out a work observation sheet.

[0086] Step 6: The production operations control team uses the shop floor production plan matrix in the knowledge base to monitor whether the production operations team is strictly adhering to work specifications and standards. They compare actual work results with the planned matrix to determine deviations, such as whether the actual paint color difference ΔE value exceeds the expected range and whether the clearcoat film thickness meets the standard. Regarding the hood paint color difference issue reported by the production operations team, the system feedback module uses a model to perform semantic image segmentation on the hood paint color difference photo to identify the area percentage of the color difference region.

[0087] Step 7: The inspection and correction module calculates the deviation matrix. According to the preset rules, the error rate exceeds the threshold of 0.1%. The system determines that manual intervention is required and generates a deviation analysis report. The system automatically generates an "Emergency Quality Incident Form" and simultaneously pushes an alarm to the mobile terminals of the paint shop director and quality engineer through WeChat for Business.

[0088] Step 8: The production operation layer formulates a correction plan and issues instructions based on the workshop production plan matrix, deviation analysis report, and knowledge base preset solution library. The correction plan is: IF humidity>80% THEN adjustment plan: 1. Heat the paint to 35℃ (originally 30℃).

[0089] 2. The atomization pressure is increased to 450kPa (originally 400kPa).

[0090] 3. Add a pre-drying process (60℃×5min).

[0091] Step 9: The production operation layer receives the instruction to carry out corrective activities, implement the corrective treatment plan, and initiate step 5, which is to report the work results and problems of the corrective activities to the production operation control layer.

[0092] Step 10: The production and operations control team analyzed the anomaly and initiated problem resolution, specifically by tracing production batches and identifying vehicle painting times. By retrieving historical data, they identified fluctuations in the spray robot's parameters. Correlating this with environmental data, they ultimately determined that excessive humidity was the cause of the poor paint atomization, and generated a root cause analysis report. This deviation was caused by persistently exceeding humidity standards (85%), which increased the paint atomization particle size to 55μm (standard: 40±5μm). This, combined with robot positioning drift, resulted in abnormal optical properties of the paint film. Engineer Wang from the Equipment Department was responsible for system tracing and conducted problem resolution.

[0093] Step 11: The system records the correction process and solutions, automatically generates case reports, and stores them in the knowledge base. This knowledge base is then used to improve workflow specifications and operating standards. Specifically, a new environmental compensation rule has been added: if the weather forecast shows humidity levels above 75% for three consecutive days, a spray parameter adjustment plan will be triggered 12 hours in advance. The warning threshold is optimized, lowering the humidity monitoring warning line from 80% to 78%. A "High Humidity Environment Spraying Operation Course" has been added to the knowledge base.

[0094] It should be noted that, in this document, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0095] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A quantitative closed-loop control system for automobile production based on an active feedback mechanism, characterized by: The system comprises: The plan management module is used to formulate and break down plans, quantify tasks using a matrix-based quantification method, establish a work list P, and continuously optimize it. The indicator target planning module is used to establish an evaluation indicator system M with weights based on the work list and build a multi-dimensional target matrix with embedded time schedule , quantify work content through matrix indicator goals; Execution assistance module, used to dynamically push work specifications and operating standards based on the knowledge base; Active feedback module, used to collect operation indicator data and abnormal problem information in real time and report the data; A process monitoring module to verify process compliance through inspections and trigger corrective actions; The inspection and correction module is used to calculate the comprehensive deviation value based on the deviation analysis matrix C and trigger the correction strategy or plan revision according to the threshold.

2. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 1 is characterized in that: The work list in the plan management module , where k is the number of items, is the i-th item.

3. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 1 is characterized in that: The indicator target planning module includes: The evaluation index system construction unit is established by using vectorized weight distribution method , where n is the number of indicators, is the i-th indicator; Target benchmark generation unit, constructing a two-dimensional target matrix including the time dimension , where t is the number of all time points considered, n is the total number of indicators, represents the target value of the jth indicator at the i-th time node; Comprehensive evaluation calculation unit, execution formula: , where n is the number of indicators, is the weight of the jth indicator at the i-th time node, Indicates the target value of the jth indicator at the i-th time node.

4. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 1 is characterized in that: The inspection and correction module includes: Deviation analysis unit, the deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ； Where t is the number of time points, n is the number of indicators, It is the difference between the actual value and target value of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ； in, is the weight of the jth indicator at the i-th time node; Threshold determination unit, control threshold ratio is , using three-layer judgment logic: the calculation formula is as follows: ； in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ； when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted.

5. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 1 is characterized in that: The active feedback mechanism in the active feedback module is implemented by setting an automatic triggering reporting protocol for abnormal data, and automatically generating a feedback report containing a deviation data packet and associated responsible parties when it is detected that the indicator fluctuation exceeds a preset threshold parameter.

6. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 1 is characterized in that: The system further comprises: The problem-solving module is used to locate the root cause of the problem through responsibility tracing and to track the solution process in a closed loop; The knowledge base module is used to store and iteratively update the standard knowledge graph including the solution library and specification library.

7. The automobile production quantitative closed-loop control system based on active feedback mechanism according to claim 6 is characterized in that: The optimization mechanism of the knowledge base module includes: Build a case-solution mapping model and use similarity matching algorithms to associate historical issues with solutions; Set knowledge iteration rules. When the same problem occurs more than 3 times N times, the standard specification revision process is automatically triggered.

8. A quantitative closed-loop control method for automobile production based on an active feedback mechanism, characterized in that: The method comprises: Formulate and break down plans, quantify tasks using matrix-based quantitative methods, establish a task list P, and continuously optimize it; Establish an evaluation index system M with weights based on the work list, and build a multi-dimensional target matrix with embedded time schedule , quantify work content through matrix indicator goals; Dynamically push work specifications and operating standards based on the knowledge base; Collect operational indicator data and abnormal problem information in real time and report the data; Verify process compliance through inspections and trigger corrective actions; The comprehensive deviation value is calculated based on the deviation analysis matrix C, and the correction strategy or plan revision is triggered according to the threshold.

9. The automobile production quantitative closed-loop control method based on active feedback mechanism according to claim 8 is characterized in that: The evaluation index system M including weights is established based on the work list, and a multi-dimensional target matrix with embedded time schedule is constructed. , quantifying the work content through matrix indicators and targets includes the following steps: Step 2.1: Determine the work priorities based on the work items; Step 2.2: Set up a work performance evaluation index system. The evaluation index system M is a vector containing multiple indicators, where each indicator has a different weight. The evaluation index system M is represented by a vector: ； Where n is the number of indicators, is the i-th indicator; Step 2.3: Generate a work item plan based on the work item list and evaluation indicator system; Step 2.4: Determine the multidimensional target matrix of expected results based on the evaluation indicator system and time schedule. The multidimensional target matrix O is composed of the target values of indicators based on time points and is represented by a vector: ； Where t is the number of time points, n is the number of indicators, is the target value of the jth indicator at the i-th time node; Comprehensive evaluation objectives The calculation formula is as follows: ； Where n is the number of indicators, is the weight of the jth indicator at the i-th time node; Step 2.5: Generate a work plan based on indicators and goals.

10. The automobile production quantitative closed-loop control method based on active feedback mechanism according to claim 8, characterized in that: The method of calculating the comprehensive deviation value based on the deviation analysis matrix C and triggering the deviation correction strategy or plan revision according to the threshold value includes the following steps: Step 6.1: Check the deviation between the work results and the expected goals; Step 6.2: Generate a deviation analysis report, calculate the deviation analysis matrix, and evaluate whether the deviation exceeds the preset threshold. The deviation analysis matrix C is based on the target difference between the time point and the indicator, and is represented by a vector: ； Where t is the number of time points, n is the number of indicators, is the target difference of the jth indicator at the i-th time node, the comprehensive deviation value The calculation formula is as follows: ； in, is the weight of the jth indicator at the i-th time node; Step 6.3: If there is a deviation but it does not exceed the control threshold, a correction activity is triggered and the corresponding treatment plan is executed according to the preset solution library. If there is a deviation but it exceeds the control threshold, a plan modification activity is triggered to adjust the work content plan. If there is no deviation, the process ends. This rule is expressed as the following formula: The control threshold ratio is , the calculation formula is as follows: ； in, is the maximum value, is the minimum value, is the standard value; Corrective activity flag variable The values of are as follows: ； when When the value is 0, the process ends. When the value is 1, the correction activity is triggered. When the value is -1, the work content plan is adjusted; Step 6.4: Develop a correction plan based on the solution library preset in the knowledge base; Step 6.5: Issue correction instructions; Step 6.6: Receive correction instructions and carry out correction activities.

Images