Quality management system and method based on digitization

Abstract

The invention discloses a digitalization-based quality management system and method. The system comprises a data processing module, a data acquisition module and a quality management module, relates to the technical field of quality management, and solves the technical problem that a quality management system in the prior art is not high in accuracy and reliability during operation. According to the invention, the product yield of each process link is obtained based on the quality inspection data; calculating a quality evaluation coefficient of the production line based on the product yield and the production data; inputting the quality evaluation coefficient into a quality prediction model to obtain a quality evaluation coefficient prediction value; and adjusting the process links in the production line based on the quality evaluation coefficient predicted value and the process influence coefficient. According to the method and the system, the entry durations of the workers in each process link are acquired, the workers are marked as skilled workers and non-skilled workers according to the entry durations, and the quantity of the skilled workers and the quantity of the non-skilled workers are substituted into a formula to calculate the corresponding proportionality coefficient, so that the operation accuracy and reliability of the quality management system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of quality management, and in particular relates to a quality management system and method based on digitization. Background Art

[0002] With the rapid development of science and technology, traditional quality management methods can no longer meet the needs of modern enterprises for efficient and accurate management. Traditional methods rely heavily on manual operations and paper documents, which are not only inefficient but also prone to errors. In recent years, digital technology has been widely used in various fields, bringing innovation to quality management. Digital technology can realize the automated collection, analysis and visualization of data, and improve management efficiency and accuracy. Therefore, a digital-based quality management system and method is invented to solve the shortcomings of traditional quality management methods and enhance corporate competitiveness and product quality.

[0003] The prior art achieves quality management of the production line to a certain extent by performing quality inspection on the products of each process in the production line to obtain the qualified rate of the products of each process; however, the operating conditions of the equipment in different process links in the production line are not the same. For example, the operating years, operating load and maintenance times will affect the quality of the products produced by the production line. The prior art solution accurately evaluates the degree of influence of the operating conditions of the equipment in different process links in the production line on the quality of the production line, resulting in low reliability of quality management of the products in the production line; in addition, the proficiency of the workers in the production line is highly correlated with the quality of the products. The prior art makes it difficult to evaluate the production quality of the production line based on the proficiency of the workers, resulting in low accuracy and reliability of the quality management system during operation.

[0004] The present invention provides a digital quality management system and method to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a quality management system and method based on digitalization, which is used to solve the problem that the equipment operating conditions in different process links in the production line are not the same. The prior art solution accurately evaluates the impact of the equipment operating conditions in different process links in the production line on the quality of the production line, resulting in low reliability of quality management of products in the production line; in addition, the proficiency of workers in the production line is highly correlated with the quality of the products. The prior art makes it difficult to evaluate the production quality of the production line based on the proficiency of the workers, resulting in technical problems such as low accuracy and reliability of the quality management system during operation.

[0006] To achieve the above-mentioned object, the first aspect of the present invention provides a digital quality management system, comprising: a data processing module, and a data acquisition module and a quality management module connected thereto;

[0007] The data acquisition module is used to collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles;

[0008] The data processing module is used to obtain the product yield of each process link based on the quality inspection data; calculate the quality assessment coefficient of the production line based on the product yield and production data; input the quality assessment coefficient into the quality prediction model to obtain the predicted value of the quality assessment coefficient; wherein the quality prediction model is constructed based on the artificial intelligence model;

[0009] The quality management module is used to adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

[0010] Preferably, the method of obtaining the product yield of each process step based on the quality inspection data includes:

[0011] Extract the quality inspection data of each process link in several consecutive cycles, mark the process links as i in turn, and obtain the qualified product quantity curve fi(x) and defective product quantity curve gi(x) of process link i through curve fitting; Calculate the total number of qualified products HPSi in process link i; by formula Calculate the total number of defective products CPSi in process step i; by the formula Calculate the product yield CPLi of process link i; the quality inspection data includes the number of qualified products and the number of defective products; t is time, t∈[t1,t2], t1 is the start time of several consecutive cycles, and t2 is the end time of several consecutive cycles; i=1, 2,…, n, n is the total number of process links.

[0012] Preferably, the calculation of the quality assessment coefficient of the production line based on the product yield and production data includes:

[0013] A1: Extract product yield and production data of each process link; production data includes employee information and equipment data. Employee information includes the number of workers and length of employment. Equipment data includes equipment age, equipment maintenance times, and equipment workload.

[0014] A2: Determine the corresponding process impact coefficient based on the product yield of each process link;

[0015] A3: By formula Calculate the quality assessment coefficient ZPX of the production line; where GXYi is the process influence coefficient of process link i, SBSi is the number of equipment in process link i, and GRSi is the number of workers in process link i; ai and bi are both proportional coefficients greater than 0.

[0016] Preferably, the determining of the corresponding process influence coefficient based on the product yield of each process step includes:

[0017] The product yield of each process link in several consecutive cycles is extracted, and the product yield change function of each process link is obtained through linear fitting; the first-order derivative function of the product yield change function is calculated to obtain the yield derivative function; the maximum function value of the yield derivative function is subtracted from the minimum function value to obtain the process influence coefficient corresponding to each process link.

[0018] Preferably, the product yield variation function of each process step is obtained by linear fitting, including:

[0019] The product yield of each process link in several consecutive cycles is extracted, with time as the independent variable and product yield as the dependent variable, and the product yield change curve of each process link is obtained through linear fitting.

[0020] Preferably, the method for obtaining the proportional coefficient ai includes:

[0021] Obtain the equipment age, equipment maintenance times and equipment workload from the equipment data of each process step, and use the formula Calculate the proportionality coefficient ai; where SSNi is the service life of the equipment in process link i, SWCi is the number of equipment maintenance times in process link i, and SFZi is the equipment workload in process link i.

[0022] Preferably, the method for obtaining the proportional coefficient bi includes:

[0023] B1: Obtain the number of workers and their employment time in each process;

[0024] B2: Determine whether the length of employment of each worker in each process link is more than 6 months; if yes, mark the corresponding worker as a skilled worker; if no, mark the corresponding worker as an unskilled worker;

[0025] B3: The proportionality coefficient bi is calculated by the formula bi=0.6×ln(1+SLGi)+0.4×ln(1+FLGi); where SLGi is the number of skilled workers in process link i, FLGi is the number of unskilled workers in process link i, and ln() is a logarithmic function with a natural constant as the base.

[0026] Preferably, the quality prediction model is constructed based on an artificial intelligence model, including:

[0027] Extract quality assessment coefficients of several consecutive cycles and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain a quality prediction model whose input is the quality assessment coefficients of the latest several consecutive cycles of each enterprise and whose output is the predicted value of the quality assessment coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0028] Preferably, the adjusting of the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient includes:

[0029] T1: Extract the predicted value of the quality assessment coefficient of the production line and the process influence coefficient of each process link in the production line;

[0030] T2: Sort the process influence coefficients of each process link from large to small to obtain the process troubleshooting priority sequence;

[0031] T3: Determine whether the predicted value of the quality assessment coefficient is greater than the preset quality assessment threshold; if yes, mark the quality level as first-level quality and jump to T1; if no, mark the quality level as second-level quality and jump to T4;

[0032] T4: Check and adjust each process link in the production line according to the process check priority sequence.

[0033] A second aspect of the present invention provides a quality management method based on digitalization, comprising:

[0034] S1: Collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles;

[0035] S2: Obtain the product yield of each process based on quality inspection data;

[0036] S3: Calculate the quality assessment coefficient of the production line based on product yield and production data;

[0037] S4: inputting the quality assessment coefficient into the quality prediction model to obtain a predicted value of the quality assessment coefficient;

[0038] S5: Adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention obtains the equipment service life, equipment maintenance times and equipment workload from the equipment data of each process link, and calculates the proportional coefficient ai through a formula; the present invention fully considers the influence of the equipment service life, equipment maintenance times and equipment workload of each process link in the production line on the quality of the product, so that the proportional coefficient ai is used for the subsequent calculated quality assessment coefficient to be more accurate, which is beneficial to improve the accuracy and reliability of the operation of the quality management system.

[0041] 2. The present invention obtains the length of time that workers in each process link of the production line have been employed, and marks the workers as skilled and unskilled workers according to the length of time; substitutes the number of skilled workers and the number of unskilled workers into the formula to calculate and obtain the proportionality coefficient bi; since the proficiency of workers is highly correlated with the quality of products, the present invention fully considers the impact of the number of skilled and unskilled workers in each process link on product quality, so that the calculated proportionality coefficient bi is more accurate in calculating the quality assessment coefficient of the production line, which is beneficial to improving the accuracy and reliability of the operation of the quality management system.

[0042] 3. The present invention extracts the predicted value of the quality assessment coefficient of the production line and the process influence coefficient of each process link, and sorts the troubleshooting order of each process link according to the size of the process influence coefficient to form a process troubleshooting priority sequence, which is conducive to quickly troubleshooting and locating quality problems in the production line. In addition, the present invention obtains the quality level of the production line based on the comparison between the quality assessment coefficient and the preset quality assessment threshold, and conducts targeted troubleshooting and adjustment based on the process troubleshooting priority sequence, which significantly improves production efficiency and quality level, thereby helping to improve the reliability and stability of the quality management system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is an overall flow chart of the digital quality management method of the present invention;

[0045] Figure 2 It is a principle schematic diagram of the digital quality management system of the present invention;

[0046] Figure 3 The present invention is a flow chart for adjusting the process links in the production line. DETAILED DESCRIPTION

[0047] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1-Figure 3 , the first aspect of the present invention provides a digital quality management system, including: a data processing module, and a data acquisition module and a quality management module connected thereto;

[0049] Data collection module: used to collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles;

[0050] Data processing module: used to obtain the product yield of each process link based on quality inspection data; calculate the quality assessment coefficient of the production line based on the product yield and production data; input the quality assessment coefficient into the quality prediction model to obtain the predicted value of the quality assessment coefficient; wherein the quality prediction model is constructed based on the artificial intelligence model;

[0051] Quality management module: used to adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

[0052] In this embodiment, the product yield of each process link is obtained based on the quality inspection data, including:

[0053] Extract the quality inspection data of each process link in several consecutive cycles, mark the process links as i in turn, and obtain the qualified product quantity curve fi(x) and defective product quantity curve gi(x) of process link i through curve fitting; Calculate the total number of qualified products HPSi in process link i; by formula Calculate the total number of defective products CPSi in process step i; by the formula Calculate the product yield CPLi of process link i; the quality inspection data includes the number of qualified products and the number of defective products; t is time, t∈[t1,t2], t1 is the start time of several consecutive cycles, and t2 is the end time of several consecutive cycles; i=1, 2,…, n, n is the total number of process links.

[0054] Exemplarily, a total of 3 process steps are set in the production line. In several consecutive cycles, the total number of qualified products in process step 1 is HPS1=73, and the total number of defective products is CPS1=5, the total number of qualified products in process step 2 is HPS2=67, and the total number of defective products is CPS2=2, the total number of qualified products in process step 3 is HPS3=65, and the total number of defective products is CPS3=6; the product yield of process step 1 is CPL1=93.59%, the product yield of process step 2 is CPL2=97.10%, and the product yield of process step 3 is CPL3=91.55%.

[0055] In this embodiment, the quality assessment coefficient of the production line is calculated based on the product yield and production data, including:

[0056] A1: Extract product yield and production data of each process link; production data includes employee information and equipment data. Employee information includes the number of workers and length of employment. Equipment data includes equipment age, equipment maintenance times, and equipment workload.

[0057] A2: Determine the corresponding process impact coefficient based on the product yield of each process link;

[0058] A3: By formula Calculate the quality assessment coefficient ZPX of the production line; where GXYi is the process influence coefficient of process link i, SBSi is the number of equipment in process link i, and GRSi is the number of workers in process link i; ai and bi are both proportional coefficients greater than 0.

[0059] Exemplarily, the proportional coefficients are set as a1=1.38, b1=1.63, a2=1.2, b2=1.5, a3=1.4, b3=1.7; the number of workers in process link 1 is GRS1=8, the number of equipment is SBS1=5, and the process influence coefficient is GXY1=1.7; the number of workers in process link 2 is GRS2=10, the number of equipment is SBS2=7, and the process influence coefficient is GXY2=1.8; the number of workers in process link 3 is GRS3=7, the number of equipment is SBS3=5, and the process influence coefficient is GXY3=1.4; the quality assessment coefficient ZPX≈12.17 of the production line is calculated by the formula.

[0060] In this embodiment, the corresponding process influence coefficient is determined based on the product yield of each process link, including:

[0061] The product yield of each process link in several consecutive cycles is extracted, and the product yield change function of each process link is obtained through linear fitting; the first-order derivative function of the product yield change function is calculated to obtain the yield derivative function; the maximum function value of the yield derivative function is subtracted from the minimum function value to obtain the process influence coefficient corresponding to each process link.

[0062] Exemplarily, the maximum function value of the yield derivative function of process link 1 is set to 2.2, and the minimum function value is set to 0.5, then the corresponding process influence coefficient GXY1=1.7.

[0063] The present invention obtains the corresponding product yield change function by performing linear fitting on the product yield of each process link in several continuous cycles; deriving the product yield change function to obtain the yield derivative function; subtracting the minimum function value from the maximum function value of the yield derivative function to obtain the process influence coefficient, so that the calculated process influence coefficient can accurately reflect the product yield fluctuation of each process link, which is convenient for the subsequent calculation of the quality assessment coefficient of the production line to be more accurate, thereby helping to improve the accuracy and reliability of the operation of the quality management system.

[0064] In this embodiment, the product yield change function of each process link is obtained by linear fitting, including:

[0065] The product yield of each process link in several consecutive cycles is extracted, with time as the independent variable and product yield as the dependent variable, and the product yield change curve of each process link is obtained through linear fitting.

[0066] In this embodiment, the method for obtaining the proportional coefficient ai includes:

[0067] Obtain the equipment age, equipment maintenance times and equipment workload from the equipment data of each process step, and use the formula Calculate the proportionality coefficient ai; where SSNi is the service life of the equipment in process link i, SWCi is the number of equipment maintenance times in process link i, and SFZi is the equipment workload in process link i.

[0068] Exemplarily, it is set that the equipment service life SSN1 of process step 1 is 5 years, the equipment maintenance times SWCi of process step i is 10 times, and the equipment workload SFZi of process step i is 90%; the proportional coefficient a≈1.38 is obtained by calculation through the formula.

[0069] The present invention obtains the equipment service life, equipment maintenance times and equipment workload from the equipment data of each process step, and calculates the proportionality coefficient ai through a formula; the present invention fully considers the influence of the equipment service life, equipment maintenance times and equipment workload on the quality of the product, so that the proportionality coefficient ai is used for the subsequent calculated quality assessment coefficient to be more accurate, which is beneficial to improve the accuracy and reliability of the operation of the quality management system.

[0070] In this embodiment, the method for obtaining the proportional coefficient bi includes:

[0071] B1: Obtain the number of workers and their employment time in each process;

[0072] B2: Determine whether the length of employment of each worker in each process link is more than 6 months; if yes, mark the corresponding worker as a skilled worker; if no, mark the corresponding worker as an unskilled worker;

[0073] B3: The proportionality coefficient bi is calculated by the formula bi=0.6×ln(1+SLGi)+0.4×ln(1+FLGi); where SLGi is the number of skilled workers in process link i, FLGi is the number of unskilled workers in process link i, and ln() is a logarithmic function with a natural constant as the base.

[0074] Exemplarily, the number of skilled workers in process step 1 SLG1=5 and the number of unskilled workers in process step 1 FLG1=3 are set; the proportional coefficient b≈1.63 is calculated by the formula.

[0075] The present invention obtains the length of employment of workers in each process link in the production line, determines whether the length of employment is greater than 6 months, and marks the corresponding workers as skilled workers and unskilled workers according to the determination result; substitutes the number of skilled workers and the number of unskilled workers into the formula to calculate and obtain the proportionality coefficient bi; since the proficiency of workers is highly correlated with the quality of products, the present invention fully considers the impact of the number of skilled workers and unskilled workers in each process link on product quality, so that the calculated proportionality coefficient bi is more accurate in calculating the quality assessment coefficient of the production line, which is beneficial to improving the accuracy and reliability of the operation of the quality management system.

[0076] In this embodiment, the quality prediction model is constructed based on an artificial intelligence model, including:

[0077] The quality assessment coefficients of several consecutive cycles are extracted and integrated into several groups of original data, 80% of the original data are used as training data and 20% as test data; the artificial intelligence model is trained using the training data; the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test results; finally, a quality prediction model is obtained, whose input is the quality assessment coefficients of the latest several consecutive cycles of each enterprise, and whose output is the predicted value of the quality assessment coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0078] The present invention extracts quality assessment coefficients of several consecutive cycles for training an artificial intelligence model, marks the trained artificial intelligence model as a quality prediction model, and inputs the quality assessment coefficients of several consecutive cycles into the quality prediction model to obtain predicted values ​​of the quality assessment coefficients of the prediction cycle, so that the quality management system can analyze the product production quality of the production line in advance, thereby timely checking and adjusting each process link in the production line, which is beneficial to improving the stability and reliability of the operation of the quality management system.

[0079] In this embodiment, the process links in the production line are adjusted based on the predicted value of the quality assessment coefficient and the process influence coefficient, including:

[0080] T1: Extract the predicted value of the quality assessment coefficient of the production line and the process influence coefficient of each process link in the production line;

[0081] T2: Sort the process influence coefficients of each process link from large to small to obtain the process troubleshooting priority sequence;

[0082] T3: Determine whether the predicted value of the quality assessment coefficient is greater than the preset quality assessment threshold; if yes, mark the quality level as first-level quality and jump to T1; if no, mark the quality level as second-level quality and jump to T4;

[0083] T4: Check and adjust each process link in the production line according to the process check priority sequence.

[0084] For example, the preset quality assessment threshold is set to 15; the quality assessment coefficient ZPX of the production line 预测 =13, the process influence coefficient of process link 1 GXY1 = 1.7, the process influence coefficient of process link 2 GXY2 = 1.8, and the process influence coefficient of process link 3 GXY3 = 1.4; sort the process influence coefficients of each process link in descending order to obtain the process screening priority sequence {process link 2, process link 1, process link 3}; due to the quality assessment coefficient ZPX 预测 =13 is less than the preset quality assessment threshold of 15, so the quality level of the production line is marked as level 2 quality, and jump to T4; according to the process check priority sequence {process link 2, process link 1, process link 3}, each process link in the production line is checked and adjusted.

[0085] The present invention extracts the predicted value of the quality assessment coefficient of the production line and the process influence coefficient of each process link, and sorts the troubleshooting order of each process link according to the size of the process influence coefficient to form a process troubleshooting priority sequence, which is conducive to quickly troubleshooting and locating quality problems in the production line. According to the comparison between the quality assessment coefficient and the preset quality assessment threshold, the quality level of the production line is obtained, and targeted troubleshooting and adjustment are carried out according to the process troubleshooting priority sequence, which significantly improves the production efficiency and quality level, thereby helping to improve the reliability and stability of the operation of the quality management system.

[0086] A second aspect of the present invention provides a digital quality management method, including:

[0087] S1: Collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles;

[0088] S2: Obtain the product yield of each process based on quality inspection data;

[0089] S3: Calculate the quality assessment coefficient of the production line based on product yield and production data;

[0090] S4: inputting the quality assessment coefficient into the quality prediction model to obtain a predicted value of the quality assessment coefficient;

[0091] S5: Adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

[0092] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0093] Working principle of the present invention:

[0094] The present invention collects quality inspection data and production data of products to be inspected in each process link of the production line in several continuous cycles; obtains the product yield of each process link based on the quality inspection data; calculates the quality assessment coefficient of the production line based on the product yield and the production data; inputs the quality assessment coefficient into a quality prediction model to obtain a predicted value of the quality assessment coefficient; and adjusts the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

[0095] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A digital quality management system, comprising: The data processing module, and the data acquisition module and quality management module connected thereto are characterized in that: The data acquisition module is used to collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles; The data processing module is used to obtain the product yield of each process link based on the quality inspection data; calculate the quality assessment coefficient of the production line based on the product yield and production data; input the quality assessment coefficient into the quality prediction model to obtain the predicted value of the quality assessment coefficient; wherein the quality prediction model is constructed based on the artificial intelligence model; The quality management module is used to adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

2. A digital quality management system according to claim 1, characterized in that: The method of obtaining the product yield of each process step based on the quality inspection data includes: Extract the quality inspection data of each process link in several consecutive cycles, mark the process links as i in turn, and obtain the qualified product quantity curve fi(x) and defective product quantity curve gi(x) of process link i through curve fitting; Calculate the total number of qualified products HPSi in process link i; by formula Calculate the total number of defective products CPSi in process step i; by the formula Calculate the product yield CPLi of process link i; the quality inspection data includes the number of qualified products and the number of defective products; t is time, t∈[t1,t2], t1 is the start time of several consecutive cycles, and t2 is the end time of several consecutive cycles; i=1, 2,…, n, n is the total number of process links.

3. A digital quality management system according to claim 1, characterized in that: The quality assessment coefficient of the production line is calculated based on the product yield and production data, including: A1: Extract product yield and production data of each process link; production data includes employee information and equipment data. Employee information includes the number of workers and length of employment. Equipment data includes equipment age, equipment maintenance times, and equipment workload. A2: Determine the corresponding process impact coefficient based on the product yield of each process link; A3: By formula Calculate the quality assessment coefficient ZPX of the production line; where GXYi is the process influence coefficient of process link i, SBSi is the number of equipment in process link i, and GRSi is the number of workers in process link i; ai and bi are both proportional coefficients greater than 0.

4. A digital quality management system according to claim 3, characterized in that: The corresponding process influence coefficient is determined based on the product yield of each process link, including: The product yield of each process link in several consecutive cycles is extracted, and the product yield change function of each process link is obtained through linear fitting; the first-order derivative function of the product yield change function is calculated to obtain the yield derivative function; the maximum function value of the yield derivative function is subtracted from the minimum function value to obtain the process influence coefficient corresponding to each process link.

5. A digital quality management system according to claim 4, characterized in that: The product yield change function of each process step is obtained by linear fitting, including: The product yield of each process link in several consecutive cycles is extracted, with time as the independent variable and product yield as the dependent variable, and the product yield change curve of each process link is obtained through linear fitting.

6. A digital quality management system according to claim 3, characterized in that: The method for obtaining the proportional coefficient ai includes: Obtain the equipment age, equipment maintenance times and equipment workload from the equipment data of each process step, and use the formula Calculate the proportionality coefficient ai; where SSNi is the service life of the equipment in process link i, SWCi is the number of equipment maintenance times in process link i, and SFZi is the equipment workload in process link i.

7. A digital quality management system according to claim 3, characterized in that: The method for obtaining the proportional coefficient bi includes: B1: Obtain the number of workers and their employment time in each process; B2: Determine whether the length of employment of each worker in each process link is more than 6 months; if yes, mark the corresponding worker as a skilled worker; if no, mark the corresponding worker as an unskilled worker; B3: The proportionality coefficient bi is calculated by the formula bi=0.6×ln(1+SLGi)+0.4×ln(1+FLGi); where SLGi is the number of skilled workers in process link i, FLGi is the number of unskilled workers in process link i, and ln() is a logarithmic function with a natural constant as the base.

8. A digital quality management system according to claim 1, characterized in that: The quality prediction model is constructed based on an artificial intelligence model, including: Extract quality assessment coefficients of several consecutive cycles and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain a quality prediction model whose input is the quality assessment coefficients of the latest several consecutive cycles of each enterprise and whose output is the predicted value of the quality assessment coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

9. A digital quality management system according to claim 1, characterized in that: The step of adjusting the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient includes: T1: Extract the predicted value of the quality assessment coefficient of the production line and the process influence coefficient of each process link in the production line; T2: Sort the process influence coefficients of each process link from large to small to obtain the process troubleshooting priority sequence; T3: Determine whether the predicted value of the quality assessment coefficient is greater than the preset quality assessment threshold; if yes, mark the quality level as first-level quality and jump to T1; if no, mark the quality level as second-level quality and jump to T4; T4: Check and adjust each process link in the production line according to the process check priority sequence.

10. A digital quality management method, based on the digital quality management system according to any one of claims 1 to 9, characterized in that: include: S1: Collect the quality inspection data and production data of the products to be inspected in each process link of the production line in several consecutive cycles; S2: Obtain the product yield of each process based on quality inspection data; S3: Calculate the quality assessment coefficient of the production line based on product yield and production data; S4: inputting the quality assessment coefficient into the quality prediction model to obtain a predicted value of the quality assessment coefficient; S5: Adjust the process links in the production line based on the predicted value of the quality assessment coefficient and the process influence coefficient.

Images