Process quality control method combining FMEA analysis and multivariate control chart

Abstract

The invention discloses a process quality control method combining FMEA analysis and a multivariate control chart. The method comprises the following steps: planning and preparing PFMEA; identifying special quality indexes of the process parameter diagram; a control limit and an abnormal mode are determined through failure analysis and risk analysis in the PFMEA, so that abnormal fluctuation in the product machining process is accurately identified; performing multivariate quality index decoupling and correlation analysis; constructing a multivariate SPC control chart; performing abnormal mode identification and source tracing on the multivariate SPC control chart; pFMEA optimization is combined with an SPC control chart to adjust the sampling frequency and the control limit. According to the method, the technical problems that the product quality is difficult to control through frequent detection during large part machining, a single-variable SPC control chart is inaccurate in abnormal mode recognition, the SPC control chart cannot be formulated according to the actual production process, and the production process cannot be effectively improved by the SPC technology are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of product processing quality control, and in particular to a process quality control method combining FMEA analysis and multivariate control charts. Background Art

[0002] With the rapid development of the world economy, quality management has played an increasingly important role in the manufacturing industry. At present, for small batches and multiple varieties of parts, the processing quality of a single process or a single piece of equipment cannot be controlled in a timely and effective manner, so its quality management is different from the quality management strategy implemented in the general manufacturing industry. The accuracy of large processing machine tools is inconvenient to monitor and control, and the precision quality inspection of large parts cannot be carried out at high frequencies. The use of large castings for trial processing is likely to increase manufacturing costs. Therefore, process capability assessment and real-time processing status monitoring for such manufacturing processes are of great significance for controlling enterprise quality costs.

[0003] Failure mode and effects analysis (FMEA) and statistical process control (SPC) technology are standardized process quality analysis tools in the IATF 16949 quality management system. In the implementation verification of many foreign companies, SPC has been used as an important and effective monitoring method. By using SPC technology, key process parameters, daily equipment status and product production quality in the production process are monitored. However, before using SPC technology for analysis, the monitoring object, control degree and sample space are often determined by subjective judgment; in addition, the single variable SPC control chart also has problems such as inaccurate identification of abnormal patterns, SPC control charts cannot be formulated according to the actual production process, and SPC technology cannot effectively improve the production process. Using the corresponding analysis stage in the process failure mode and effects analysis (PFMEA), a reference that is more in line with actual production can be provided for the formulation of SPC. Regularly evaluate and review FMEA and SPC at the same time to ensure that the two are coordinated and work together to improve process quality and reliability. Summary of the invention

[0004] In order to overcome the shortcomings of the above technologies, the present invention provides a process quality control method combining FMEA analysis and multivariate control charts. The method solves the technical problems that it is difficult to control product quality through frequent testing in large-scale parts processing, the single variable SPC control chart is inaccurate in identifying abnormal patterns, the SPC control chart cannot be formulated according to the actual production process, and the SPC technology cannot effectively improve the production process.

[0005] Terminology explanation:

[0006] 1. FMEA: Failure Mode and Effect Analysis.

[0007] 2. DFMEA: Design Failure Mode and Effect Analysis.

[0008] 3. PFMEA: Process Failure Mode and Effect Analysis.

[0009] 4. SPC: Statistical Process Control.

[0010] 5. BOM: Bill of Material.

[0011] The technical solution adopted by the present invention to overcome the technical problems is:

[0012] A process quality control method combining FMEA analysis and multivariate control charts, comprising the following steps:

[0013] S1. The manufacturer shall plan and prepare PFMEA according to internal and external requirements and technical standards and at least in combination with the graphic technical documents, DFMEA and bill of materials;

[0014] S2. Identification of special quality indicators in process parameter diagram: Develop process parameter diagram for parts processing through structural analysis and functional analysis of PFMEA, and then identify special quality indicators in process parameter diagram as the key objects of SPC monitoring, and determine the major issues and special process characteristics that need to be recorded in the product processing process;

[0015] S3. Determine control limits and abnormal modes through failure analysis and risk analysis in PFMEA, so as to accurately identify abnormal fluctuations in product processing;

[0016] S4. Decoupling and correlation analysis of multiple quality indicators: For the special quality indicators that need to be monitored in step S2, correlation analysis and principal component analysis are used to perform correlation analysis on the features from the perspective of data analysis;

[0017] S5. Construct a multivariate SPC control chart: Use the different special quality indicator measurement data determined in step S2 and step S4 to perform visual analysis, perform data transformation on the different quality characteristic measurement data, so that they can be displayed in the SPC control chart with the same center limit and control limit, and mark the abnormal fluctuations of major events in the product processing process at the corresponding time nodes;

[0018] S6. Identification and tracing of abnormal patterns of multivariate SPC control charts: For the multivariate SPC control chart obtained in step S5, identify abnormalities in quality data through eight typical abnormality judgment criteria of the SPC control chart and joint judgment between multiple special quality indicators, and judge the faults in the product processing process by combining the abnormal pattern determined in step S3 and the abnormal fluctuations of the major events marked in step S5;

[0019] S7. PFMEA optimization combined with SPC control chart to adjust sampling frequency and control limits: After evaluating the implementation effect of preventive measures and detection measures in the PFMEA optimization stage, the sampling frequency and control limit coefficients of the SPC control chart are adjusted in combination with SPC control chart analysis.

[0020] Furthermore, in step S1, internal and external demands include external customer demands and internal customer demands, wherein external customer demands at least include consumer demands, and internal customer demands at least include downstream process demands; technical standards at least include regulatory requirements, technical requirements, and specification standards; and graphical technical documents at least include drawings, charts, and design drawings.

[0021] Furthermore, step S2 specifically includes:

[0022] S21. Develop a parts processing flow chart through PFMEA structural analysis, which is used to describe at least the structure of the overall process and sub-processes in the manufacturing process, and reflect the connection between the various steps and elements in the product processing process;

[0023] S22, formulating a process parameter diagram for part processing by using the process flow chart obtained in step S21 through functional analysis of PFMEA, wherein the process parameter diagram at least includes parameters that affect input and output process functions, and is used to display the environment in which the entire product processing process is located;

[0024] S23. Use the process parameter diagram of PFMEA, which lists at least the important influencing factors, product requirements and process control elements, to determine the special quality indicators that need to be monitored during product processing, the major issues that need to be recorded and the special process characteristics.

[0025] Furthermore, step S3 specifically includes:

[0026] S31. Determine the potential failure modes, failure effects and failure causes of each process function through failure analysis in PFMEA, thereby establishing a failure chain;

[0027] S32. According to different failure causes, make corresponding changes to the abnormal mode of the SPC control chart of the corresponding product quality characteristics;

[0028] S33. According to the results of failure analysis, preventive measures and detection measures are determined through risk analysis in PFMEA, and severity, frequency and detection rating are performed for each failure chain to determine the priority of measures in the failure chain. Preventive measures refer to measures to eliminate the occurrence of failure causes or failure modes in process design, and preventive control measures are verified during the process verification before sample and equipment acceptance and formal production. Detection measures refer to the detection of failure causes and failure modes by automatic or manual methods during or after the product leaves the production process.

[0029] S34. Select the special quality indicators that need to be monitored based on the priority of the measures, as well as the control limits and abnormal patterns of the corresponding special quality indicators, and determine the preliminary sampling frequency of product quality inspection.

[0030] Furthermore, step S31 specifically includes:

[0031] S311. Based on the actual product processing, combined with DFMEA and process parameter diagrams, list the potential failure modes of the operations specified in the product production process;

[0032] S312, determining the failure impact of the failure mode at least through a processing flow chart and a processing process card;

[0033] S313, finding the key process influencing factors in the failure mode in the process parameter diagram, and determining the failure cause of the failure mode;

[0034] S314. According to the analysis of step S311 to step S313, an invalidation chain is established.

[0035] Furthermore, in step S33, the priority of measures is rated using three indicators SOD, where S represents the severity of the failure impact, O represents the frequency of the failure cause and failure mode, and D represents the difficulty of detecting the failure cause and failure mode.

[0036] Furthermore, step S4 specifically includes:

[0037] After failure analysis and risk analysis, several special quality indicators that need to be monitored are selected. There is overlap in the processing processes between the special quality indicators, and there is coupling between the measurement data of the special quality indicators in the time series; the correlation analysis method is used to measure and screen quality features that are higher than the preset correlation threshold. The quality features that are higher than the preset correlation threshold are monitored through one of the quality features as a representative; the principal component analysis method is used to convert related characteristics into unrelated new characteristics.

[0038] Furthermore, step S5 specifically includes:

[0039] S51, respectively collecting test data of different special quality indicators, deleting abnormal values ​​in the data, and filling in missing values;

[0040] S52, normalizing the data of different special quality indicators, calculating the deviation rate of the measurement error relative to the tolerance, and drawing a single value control chart instead of the test data, calculating the moving range of the test data and calculating the moving range mean as the data of the moving range control chart;

[0041] S53. Plot the error offsets of different special quality indicators on the same SPC single value control chart and moving range control chart. The center limit of the single value control chart is 0, the upper control limit is +A2, and the lower control limit is -A2. The center limit of the moving range control chart is 1, the upper control limit is D3, and the lower control limit is 0. Among them, A2 and D3 are control limit coefficients.

[0042] Furthermore, step S6 specifically includes:

[0043] S61, combining the abnormal pattern determined in step S3 and eight typical abnormality judgment criteria of the SPC control chart, comprehensively judging the abnormal pattern of the multivariate SPC control chart;

[0044] S62. If a single special quality indicator shows an abnormal pattern, the abnormal pattern judgment of the traditional SPC control chart is used directly; if multiple quality indicators show abnormal patterns at the same time, the relationship between the processing procedures is used for judgment;

[0045] S63. After the abnormal mode is identified, combined with the major event record table determined in step S2, it is possible to trace back to the potential failure mode or the failure cause and failure mode that have already occurred in the product processing process.

[0046] Furthermore, step S7 specifically includes:

[0047] S71. Reduce the severity level of failure effects and the frequency of failure causes by changing product design or product processing, and improve the ability to detect failure causes and failure modes by using error-proofing detection;

[0048] S72. Re-evaluate the severity, frequency, and detectability of the current failure mode and set new action priorities;

[0049] S73. When the severity and frequency of the failure mode decrease, reduce the frequency of quality inspection sampling and reduce the upper and lower control limits of the SPC control chart.

[0050] The beneficial effects of the present invention are:

[0051] The present invention generates a multivariate SPC control chart based on PFMEA analysis, and combines PFMEA and SPC to perform process quality control. Through the analysis tool of PFMEA, key characteristics with high severity and high risk priority rating are used as special quality indicators that SPC needs to focus on monitoring, and reasonable theoretical support is provided for the determination of SPC control limits and sample spaces, so that the implementation of SPC technology is more objective, and abnormal processing states can be quickly and accurately identified; through the joint diagnosis of multivariate SPC control charts, specific faults such as abnormal modes of multiple quality indicators appearing at the same time can be more accurately identified; by regularly reviewing PFMEA output and SPC data, the process and control strategy can be further optimized according to actual conditions to achieve continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a module relationship diagram involved in an embodiment of the present invention.

[0053] Figure 2 The present invention is a flowchart of a process quality control method combining FMEA analysis and multivariate control charts according to an embodiment of the present invention.

[0054] Figure 3 A flowchart of the processing process of a head plate part constructed according to an embodiment of the present invention.

[0055] Figure 4 A process parameter diagram of a head plate part constructed according to an embodiment of the present invention.

[0056] Figure 5 It is a schematic diagram of a failure chain of a partial processing process of a head plate part described in an embodiment of the present invention.

[0057] Figure 6 This is a schematic diagram of the control diagram partitioning according to an embodiment of the present invention.

[0058] Figure 7 This is a single value control chart taking L2 and L4 as examples in the multivariate SPC control chart constructed according to an embodiment of the present invention.

[0059] Figure 8 The moving range control chart of L2 and L4 in the multivariate SPC control chart constructed according to the embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to facilitate those skilled in the art to better understand the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The following is only exemplary and does not limit the protection scope of the present invention.

[0061] The present invention discloses a process quality control method combining FMEA analysis and multivariate control chart, comprising the following steps:

[0062] S1. The manufacturer shall plan and prepare PFMEA according to internal and external requirements and technical standards and at least in combination with the graphic technical documents, DFMEA and bill of materials;

[0063] S2. Identification of special quality indicators in process parameter diagram: Develop process parameter diagram for parts processing through structural analysis and functional analysis of PFMEA, and then identify special quality indicators in process parameter diagram as the key objects of SPC monitoring, and determine the major issues and special process characteristics that need to be recorded in the product processing process;

[0064] S3. Determine control limits and abnormal modes through failure analysis and risk analysis in PFMEA, so as to accurately identify abnormal fluctuations in product processing;

[0065] S4. Decoupling and correlation analysis of multiple quality indicators: For the special quality indicators that need to be monitored in step S2, correlation analysis and principal component analysis are used to perform correlation analysis on the features from the perspective of data analysis;

[0066] S5. Construct a multivariate SPC control chart: Use the different special quality indicator measurement data determined in step S2 and step S4 to perform visual analysis, perform data transformation on the different quality characteristic measurement data, so that they can be displayed in the SPC control chart with the same center limit and control limit, and mark the abnormal fluctuations of major events in the product processing process at the corresponding time nodes;

[0067] S6. Identification and tracing of abnormal patterns of multivariate SPC control charts: For the multivariate SPC control chart obtained in step S5, identify abnormalities in quality data through eight typical abnormality judgment criteria of the SPC control chart and joint judgment between multiple special quality indicators, and judge the faults in the product processing process by combining the abnormal pattern determined in step S3 and the abnormal fluctuations of the major events marked in step S5;

[0068] S7. PFMEA optimization combined with SPC control chart to adjust sampling frequency and control limits: After evaluating the implementation effect of preventive measures and detection measures in the PFMEA optimization stage, the sampling frequency and control limit coefficients of the SPC control chart are adjusted in combination with SPC control chart analysis.

[0069] In order to better understand the above technical solution, the exemplary embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. These are only exemplary embodiments of the present invention. However, it should be understood that the present invention can also be implemented in various forms and is not limited to the embodiments described herein. These embodiments are intended to enable those skilled in the art to understand the present invention more clearly and thoroughly.

[0070] The process quality control method described in this embodiment combines FMEA analysis and multivariate control charts, such as Figure 1 and Figure 2 As shown, the following steps are included:

[0071] S1. The manufacturer plans and prepares PFMEA based on internal and external requirements and technical standards and at least in combination with the graphic technical documents, DFMEA, and bill of materials (BOM). The main purpose is to determine the scope of analysis, determine team responsibilities, and collect relevant information. The purpose is to lay the foundation for subsequent failure mode analysis.

[0072] Specifically, internal and external requirements include external customer requirements and internal customer requirements, where external customer requirements at least include consumer requirements, and internal customer requirements at least include downstream process requirements. Technical standards at least include regulatory requirements, technical requirements, and specification standards. Graphic technical documents at least include drawings, charts, and design drawings.

[0073] S2. Identification of special quality indicators in process parameter diagram: Develop a process parameter diagram for part processing through structural analysis and functional analysis of PFMEA. Then identify the special quality indicators in the process parameter diagram as the focus of SPC monitoring, and determine the important matters and special process characteristics that need to be recorded during product processing.

[0074] Specifically, step S2 includes:

[0075] S21. Develop a parts processing flow chart through PFMEA structural analysis, such as Figure 3 The figure below shows the process flow chart of a head plate part, which is used to describe the structure of the overall process and sub-processes in the manufacturing process, and reflects the connection between the steps and elements in the product processing process. As an auxiliary tool, the process flow chart not only helps define the process, but also provides a visual basis for the functional analysis of PFMEA.

[0076] S22, through the functional analysis of PFMEA and using the processing flow chart obtained in step S21, a process parameter diagram for part processing is formulated, such as Figure 4 The figure shows the process parameter diagram of a head plate part. Specifically, the functional analysis focuses on factors such as the manufacturing environment, cycle, occupational health and safety, and environmental impact, aiming to identify special quality indicators of products and processes. These special quality indicators are the focus of control to ensure that safety, regulatory compliance, and customer-focused functional, performance, and technical requirements are met, while optimizing costs to improve customer satisfaction. The process parameter diagram includes at least the parameters that affect the input and output process functions, and is used to show the environment in which the entire product processing process is located.

[0077] S23. Use the process parameter diagram of PFMEA, which lists at least the important influencing factors, product requirements and process control elements, to determine the special quality indicators that need to be monitored, the major issues that need to be recorded and the special process characteristics during product processing, to provide data reference for SPC's monitoring of special quality indicators, major issues that need to be recorded and special process characteristics, and to provide a basis for subsequent fault tracing.

[0078] S3. Determine control limits and abnormal modes through failure analysis and risk analysis in PFMEA, so as to accurately identify abnormal fluctuations in product processing.

[0079] Specifically, step S3 includes:

[0080] S31. Through failure analysis in PFMEA, the potential failure modes, failure effects and failure causes of each process function are determined, thereby establishing a failure chain in order to provide a basis for risk analysis.

[0081] In this embodiment, step S31 specifically includes:

[0082] S311. Based on the actual product processing, combined with DFMEA and process parameter diagrams, list the potential failure modes of the operations specified in the product production process.

[0083] S312. Determine the failure impact of the failure mode at least through a processing flow chart and a processing process card.

[0084] S313. Find the key process influencing factors in the failure mode in the process parameter diagram, and determine the failure cause of the failure mode.

[0085] S314. According to the analysis of step S311 to step S313, an invalidation chain is established.

[0086] S32. According to different failure causes, make corresponding changes to the abnormal mode of the SPC control chart of the corresponding product quality characteristics.

[0087] S33. According to the results of failure analysis, preventive measures and detection measures are determined through risk analysis in PFMEA, and severity, frequency and detection ratings are performed for each failure chain to determine the priority of the measures of the failure chain. The rating of the priority of measures uses three indicators, SOD, where S represents the severity of the failure impact, O represents the frequency of occurrence of the failure cause and failure mode, and D represents the difficulty of detecting the failure cause and failure mode. Using process failure modes and special quality indicators, the monitoring object of SPC can be determined by referring to the detection degree D rating; using failure causes and special process characteristics, the abnormal mode of SPC can be determined by referring to the frequency O rating; using failure effects and severity S ratings and frequency O ratings, the sample space of SPC, that is, the sampling frequency, can be determined. Specifically, preventive measures refer to measures to eliminate the occurrence of failure causes or failure modes in process design, and preventive control measures are verified during the process verification before sample and equipment acceptance and formal production. Detection measures refer to the detection of failure causes and failure modes by automatic or manual methods during or after the product leaves the production process.

[0088] This embodiment takes a certain processing process of a head plate part as an example to perform failure analysis and construct a failure chain of the processing process. Some failure chains are as follows: Figure 5 As shown; the partial failure risk rating of a certain processing process of the head plate part is shown in Table 1.

[0089] Table 1 Partial failure risk rating table for a certain processing process of a head plate part

[0090]

[0091] The measures in Table 1 are divided into three priority levels: high, medium, and low, represented by H, M, and L, respectively. The PFMEA team needs to formulate different measures for different measures. The key product characteristics corresponding to the failure modes with high failure severity ratings and high frequency are selected as the monitoring object space of multivariate SPC, and the relevant algorithms are designed to process the SOD rating.

[0092] After selecting the corresponding failure mode, determine the failure cause, and formulate the abnormal mode type of the multivariate SPC control chart according to the failure cause. The corresponding standards of the abnormal mode and failure cause of the multivariate SPC control chart are shown in Table 2. In Table 2, It represents the moving range within the group. Represents the mean within the group, UCL represents the upper specification limit of the SPC control chart, LSL represents the lower specification limit of the SPC control chart, and the upper and lower control limits are located at three standard deviations above and below the center line respectively. The control chart is divided into six areas, such as Figure 6As shown, the width of each zone is a standard deviation σ, and the six zones are numbered A, B, C, C, B, A from top to bottom. Two zones with the same number are symmetrical relative to the center line. The details of the specific cause of failure should be adjusted and recorded according to the on-site production process.

[0093] Table 2 Corresponding standards between abnormal modes and failure causes of multivariate SPC control charts

[0094]

[0095] S34. Select the special quality indicators that need to be monitored based on the priority of the measures, as well as the control limits and abnormal patterns of the corresponding special quality indicators, and determine the preliminary sampling frequency of product quality inspection.

[0096] S4. Decoupling and correlation analysis of multiple quality indicators: For the special quality indicators that need to be monitored in step S2, correlation analysis and principal component analysis are used to perform correlation analysis on the features from the perspective of data analysis.

[0097] Specifically, step S4 includes:

[0098] After failure analysis and risk analysis, several special quality indicators that need to be monitored are selected. There is overlap in the processing processes between the special quality indicators, and there is coupling between the measurement data of the special quality indicators in the time series; the correlation analysis method is used to measure and screen quality features that are higher than the preset correlation threshold. The quality features that are higher than the preset correlation threshold are monitored through one of the quality features as a representative; the principal component analysis method is used to convert related characteristics into unrelated new characteristics.

[0099] In the case of this embodiment, after the analysis of step S3 above, eight quality features are selected, including five dimensional features and three shape and position features. After preliminary analysis of the data features, the three shape and position features are relatively independent, so all three shape and position features need to be monitored using a multivariate SPC control chart; in this embodiment, the Pearson coefficient and the Spearman coefficient are preferably used to perform correlation analysis on the five dimensional features, and two of them are selected for monitoring.

[0100] The calculation formula of Pearson coefficient is as follows:

[0101] ρ P[X,Y] = cov(X,Y) σ X × σ Y = E((X- μ X )(Y- μ Y )) σ X × σ Y = E XY -E(X)E(Y) E X 2 - E 2 (X) × E Y 2 - E 2 (Y) (1)

[0102] In formula (1), , They represent two groups of variables for correlation analysis. ρ P[X,Y] Representation variables With variables The Pearson coefficient, represents the expected value of a set of variables, and Respectively represent variables and variables The standard deviation of Representation variables With variables The covariance between and Respectively represent variables and variables The mean of .

[0103] The calculation formula of Spearman coefficient is as follows:

[0104] (2)

[0105] In formula (2), represents the Spearman coefficient of two groups of variables, Indicates The rank difference of the data pairs is Represents the total number of observed samples.

[0106] The comprehensive correlation coefficient is calculated by the Pearson coefficient and the Spearman coefficient. The calculation formula of the comprehensive correlation coefficient is as follows:

[0107] ρ = λ × ρ P[X,Y] + 1-λ × ρ S (3)

[0108] In formula (3), Representation variables With variables The comprehensive correlation coefficient between Represents the weight coefficient between Pearson's coefficient and Spearman's coefficient.

[0109] In the case of this embodiment, since the distribution of quality characteristics does not strictly conform to the normal distribution, the Pearson coefficient and the Spearman coefficient are used to evaluate the correlation between quality characteristics. The correlation scores of the five quality indicators are shown in Table 3. L1~L5 are the five quality indicators of the part, representing five positional accuracies. L2 and L4 are selected together with the three shape and position characteristics to form the monitoring object space of the multivariate SPC control chart.

[0110] Table 3 Correlation coefficients of various quality indicators

[0111]

[0112] S5. Construct a multivariate SPC control chart: Use the different special quality indicator measurement data determined in step S2 and step S4 to perform visual analysis, transform the different quality characteristic measurement data, and display them in the SPC control chart with the same center limit and control limit, and mark the abnormal fluctuations of major events in the product processing process at the corresponding time nodes.

[0113] Specifically, step S5 includes:

[0114] S51. Collect test data of different special quality indicators respectively, delete outliers in the data, and fill in missing values.

[0115] S52. Normalize the data of different special quality indicators, calculate the deviation rate of the measurement error relative to the tolerance, and draw a single value control chart instead of the test data, calculate the moving range of the test data and calculate the moving range mean as the data of the moving range control chart.

[0116] S53. Plot the error offsets of different special quality indicators on the same SPC single value control chart and moving range control chart. The center limit of the single value control chart is 0, the upper control limit is +A2, and the lower control limit is -A2. The center limit of the moving range control chart is 1, the upper control limit is D3, and the lower control limit is 0. Among them, A2 and D3 are control limit coefficients.

[0117] Since it is not convenient to conduct large-scale sampling inspection in this embodiment, only the first and last pieces and the middle pieces of the corresponding order are inspected in actual production, so the single value-moving range control chart can be selected, that is, the single value control chart and the moving range control chart.

[0118] First, select 30 recent data sets for process capability assessment (note that data stratification selection and grouping should be reasonable), calculate the process capability index and process parameters to calculate the control limits of the control chart. Calculate the moving range of the data and unify it. After calculating the moving range of each type of data, calculate the mean of the moving range, divide all the ranges by the mean of the range of the corresponding feature, and scale the mean of the range of different features to 1. In this embodiment, the preferred upper limit D3 is 3.267. Calculate the mean of the data and unify the control limits. Subtract the mean of the corresponding feature from the data of different features, and unify the mean line to the 0 line; then divide the processed data by the corresponding mean of the range, and unify the upper and lower control limits to ±2.660 lines, that is, A2=2.660, such as Figure 7 and Figure 8 shown.

[0119] The specific calculation formula is as follows:

[0120] (4)

[0121] (5)

[0122] (6)

[0123] Formula (4)-Formula (6), Indicates The moving range between a value and the previous value, and Represents the current value and the previous value respectively. represents the mean of the moving range within the group. Represents the center line of the moving range control chart. represents the upper control limit of the moving range control chart, represents the lower control limit of the moving range chart, represents the within-group mean, represents the center line of the individual value control chart, represents the upper control limit of the individual value control chart, Represents the lower control limit for the individuals chart.

[0124] S6. Identification and tracing of abnormal patterns of multivariate SPC control charts: For the multivariate SPC control chart obtained in step S5, identify abnormalities in quality data through the eight typical abnormality judgment criteria of the SPC control chart and the joint judgment between multiple special quality indicators, and judge the faults in the product processing process by combining the abnormal pattern determined in step S3 and the abnormal fluctuations of major events marked in step S5.

[0125] Specifically, step S6 includes:

[0126] S61. Combine the abnormal pattern determined in step S3 and the eight typical abnormality judgment standards of the SPC control chart, and make a comprehensive judgment on the abnormal pattern of the multivariate SPC control chart. The eight typical abnormality judgment standards of the SPC control chart refer to the eight typical abnormality judgment standards specified in GB T4091-2000, which are one point outside the boundary, two-thirds outside the same side B, four-fifths outside the same side C, six points rising continuously, six points falling continuously, eight points without C, fourteen points rising and falling, and fifteen points within C.

[0127] In the case of this embodiment, eight typical abnormality judgment standards are adopted, and six main standards are selected according to the actual production situation, namely one point outside the boundary, two-thirds outside the same side B, four-fifths outside the same side C, six points rising continuously, six points falling continuously, and eight points without C (due to the limitations of the enterprise production situation and quality control costs, the other two standards are not applicable to this case) as the abnormality judgment standards in the case of this embodiment. According to the enterprise production situation and the implementation method of the present invention, the six abnormality judgment standards are refined. Specifically, according to the different forms of the typical abnormality judgment standards on both sides of the center line, the abnormal patterns of the specific corresponding data distribution and the corresponding failure causes are listed respectively, as shown in Table 2.

[0128] S62. If a single special quality indicator shows an abnormal pattern, the abnormal pattern judgment is directly made using the traditional SPC control chart; if multiple quality indicators show abnormal patterns at the same time, the judgment is made using the relationship between the processing procedures.

[0129] S63. After the abnormal mode is identified, combined with the major event record table determined in step S2, it is possible to trace back to the potential failure mode or the failure cause and failure mode that have already occurred in the product processing process.

[0130] S7. PFMEA optimization combined with SPC control chart to adjust sampling frequency and control limits: After evaluating the implementation effect of preventive measures and detection measures in the PFMEA optimization stage, the sampling frequency and control limit coefficients of the SPC control chart are adjusted in combination with SPC control chart analysis.

[0131] Specifically, step S7 includes:

[0132] S71. Reduce the severity level of failure effects and the frequency of failure causes by changing product design or product processing, and improve the ability to detect failure causes and failure modes by using error-proofing detection.

[0133] S72. Re-evaluate the severity, frequency, and detectability of the current failure mode and establish new priority measures.

[0134] S73. When the severity and frequency of the failure mode decrease, reduce the frequency of quality inspection sampling and reduce the upper and lower control limits of the SPC control chart.

[0135] In the case of this embodiment, three processing steps are selected for processing analysis and optimization measures are taken. The effects of the optimized measures are shown in Table 4.

[0136] Table 4 Effects of optimized measures

[0137]

[0138] In the case of this embodiment, the cause of the fault in the processing process was accurately located through the joint fault tracing of multivariate SPC and PFMEA; corresponding measures were taken in time to reduce the frequency of occurrence and failure impact of processing process faults, ensure production efficiency and improve product qualification rate; adjust the sampling frequency in the processing process quality control strategy and the control limit of the SPC control chart, shorten the quality control cycle and reduce quality costs.

[0139] The above only describes the basic principles and preferred embodiments of the present invention. Those skilled in the art may make many changes and improvements based on the above description, and these changes and improvements should fall within the protection scope of the present invention.

Claims

1. A process quality control method combining FMEA analysis and multivariate control chart, characterized in that: The steps include: S1. The manufacturer shall plan and prepare PFMEA according to internal and external requirements and technical standards and at least in combination with the graphic technical documents, DFMEA and bill of materials; S2. Identification of special quality indicators in process parameter diagram: Develop process parameter diagram for parts processing through structural analysis and functional analysis of PFMEA, and then identify special quality indicators in process parameter diagram as the key objects of SPC monitoring, and determine the major issues and special process characteristics that need to be recorded in the product processing process; S3. Determine control limits and abnormal modes through failure analysis and risk analysis in PFMEA, so as to accurately identify abnormal fluctuations in product processing; S4. Decoupling and correlation analysis of multiple quality indicators: For the special quality indicators that need to be monitored in step S2, correlation analysis and principal component analysis are used to perform correlation analysis on the features from the perspective of data analysis; S5. Construct a multivariate SPC control chart: Use the different special quality indicator measurement data determined in step S2 and step S4 to perform visual analysis, perform data transformation on the different quality characteristic measurement data, so that they can be displayed in the SPC control chart with the same center limit and control limit, and mark the abnormal fluctuations of major events in the product processing process at the corresponding time nodes; S6. Identification and tracing of abnormal patterns of multivariate SPC control charts: For the multivariate SPC control chart obtained in step S5, identify abnormalities in quality data through eight typical abnormality judgment criteria of the SPC control chart and joint judgment between multiple special quality indicators, and judge the faults in the product processing process by combining the abnormal pattern determined in step S3 and the abnormal fluctuations of the major events marked in step S5; S7. PFMEA optimization combined with SPC control chart to adjust sampling frequency and control limits: After evaluating the implementation effect of preventive measures and detection measures in the PFMEA optimization stage, the sampling frequency and control limit coefficients of the SPC control chart are adjusted in combination with SPC control chart analysis.

2. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1, characterized in that: In step S1, internal and external demands include external customer demands and internal customer demands, wherein external customer demands at least include consumer demands, and internal customer demands at least include downstream process demands; technical standards at least include regulatory requirements, technical requirements, and specification standards; and graphical technical documents at least include drawings, charts, and design drawings.

3. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1 is characterized in that: Step S2 specifically includes: S21. Develop a parts processing flow chart through PFMEA structural analysis, which is used to describe at least the structure of the overall process and sub-processes in the manufacturing process, and reflect the connection between the various steps and elements in the product processing process; S22, formulating a process parameter diagram for part processing by using the process flow chart obtained in step S21 through functional analysis of PFMEA, wherein the process parameter diagram at least includes parameters that affect input and output process functions, and is used to display the environment in which the entire product processing process is located; S23. Use the process parameter diagram of PFMEA, which lists at least the important influencing factors, product requirements and process control elements, to determine the special quality indicators that need to be monitored during product processing, the major issues that need to be recorded and the special process characteristics.

4. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1, characterized in that: Step S3 specifically includes: S31. Determine the potential failure modes, failure effects and failure causes of each process function through failure analysis in PFMEA, thereby establishing a failure chain; S32. According to different failure causes, make corresponding changes to the abnormal mode of the SPC control chart of the corresponding product quality characteristics; S33. According to the results of failure analysis, preventive measures and detection measures are determined through risk analysis in PFMEA, and severity, frequency and detection rating are performed for each failure chain to determine the priority of measures in the failure chain. Preventive measures refer to measures to eliminate the occurrence of failure causes or failure modes in process design, and preventive control measures are verified during the process verification before sample and equipment acceptance and formal production. Detection measures refer to the detection of failure causes and failure modes by automatic or manual methods during or after the product leaves the production process. S34. Select the special quality indicators that need to be monitored based on the priority of the measures, as well as the control limits and abnormal patterns of the corresponding special quality indicators, and determine the preliminary sampling frequency of product quality inspection.

5. The process quality control method combining FMEA analysis and multivariate control chart according to claim 4 is characterized in that: Step S31 specifically includes: S311. Based on the actual product processing, combined with DFMEA and process parameter diagrams, list the potential failure modes of the operations specified in the product production process; S312, determining the failure impact of the failure mode at least through a processing flow chart and a processing process card; S313, finding the key process influencing factors in the failure mode in the process parameter diagram, and determining the failure cause of the failure mode; S314. According to the analysis of step S311 to step S313, an invalidation chain is established.

6. The process quality control method combining FMEA analysis and multivariate control chart according to claim 4 is characterized in that: In step S33, the priority of measures is rated using the three indicators SOD, where S represents the severity of the failure impact, O represents the frequency of the failure cause and failure mode, and D represents the difficulty of detecting the failure cause and failure mode.

7. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1, characterized in that: Step S4 specifically includes: After failure analysis and risk analysis, several special quality indicators that need to be monitored are selected. There is overlap in the processing processes between the special quality indicators, and there is coupling between the measurement data of the special quality indicators in the time series; the correlation analysis method is used to measure and screen quality features that are higher than the preset correlation threshold. The quality features that are higher than the preset correlation threshold are monitored through one of the quality features as a representative; the principal component analysis method is used to convert related characteristics into unrelated new characteristics.

8. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1, characterized in that: Step S5 specifically includes: S51, respectively collecting test data of different special quality indicators, deleting abnormal values ​​in the data, and filling in missing values; S52, normalizing the data of different special quality indicators, calculating the deviation rate of the measurement error relative to the tolerance, and drawing a single value control chart instead of the test data, calculating the moving range of the test data and calculating the moving range mean as the data of the moving range control chart; S53. Plot the error offsets of different special quality indicators on the same SPC single value control chart and moving range control chart. The center limit of the single value control chart is 0, the upper control limit is +A2, and the lower control limit is -A2. The center limit of the moving range control chart is 1, the upper control limit is D3, and the lower control limit is 0. Among them, A2 and D3 are control limit coefficients.

9. The process quality control method combining FMEA analysis and multivariate control chart according to claim 1, characterized in that: Step S6 specifically includes: S61, combining the abnormal pattern determined in step S3 and eight typical abnormality judgment criteria of the SPC control chart, comprehensively judging the abnormal pattern of the multivariate SPC control chart; S62. If a single special quality indicator shows an abnormal pattern, the abnormal pattern judgment of the traditional SPC control chart is used directly; if multiple quality indicators show abnormal patterns at the same time, the relationship between the processing procedures is used for judgment; S63. After the abnormal mode is identified, combined with the major event record table determined in step S2, it is possible to trace back to the potential failure mode or the failure cause and failure mode that have already occurred in the product processing process.

10. The process quality control method combining FMEA analysis and multivariate control chart according to any one of claims 6 to 9, characterized in that: Step S7 specifically includes: S71. Reduce the severity level of failure effects and the frequency of failure causes by changing product design or product processing, and improve the ability to detect failure causes and failure modes by using error-proofing detection; S72. Re-evaluate the severity, frequency, and detectability of the current failure mode and set new action priorities; S73. When the severity and frequency of the failure mode decrease, reduce the frequency of quality inspection sampling and reduce the upper and lower control limits of the SPC control chart.

Images