Method, apparatus, medium and equipment for constructing quality control charts based on fuzzy numbers

Abstract

This invention discloses a method, apparatus, medium, and equipment for constructing quality control charts based on fuzzy numbers, belonging to the technical field of quality supervision information systems. The method includes: obtaining key quality characteristics corresponding to key processes in the manufacturing process, wherein these key quality characteristics are controlled by control charts; defining fuzzy numbers that satisfy conditions based on the key quality characteristics corresponding to the key processes, and determining the possible value range of these fuzzy numbers; and constructing a control chart corresponding to the uncertain quality characteristics by performing quality factor analysis on the key processes in the manufacturing process, based on the fuzzy numbers that satisfy the conditions, the possible value range of these fuzzy numbers, and the uncertain quality characteristics in the production process. The apparatus, medium, and equipment can be used to implement this method. It can accurately determine random errors in the manufacturing process and more accurately reflect the actual product quality situation.

Description

Technical Field

[0001] This invention relates to the field of quality supervision information system technology, and in particular to a method, apparatus, medium and equipment for constructing quality control charts based on fuzzy numbers. Background Technology

[0002] Currently, count-type and measure-type control charts are commonly used in production processes to analyze and control quality, thereby reflecting quality problems. For quality characteristics that fluctuate normally according to certain patterns during production, traditional count-type quality control charts can provide effective monitoring and control. However, modern complex equipment manufacturing is characterized by complexity, nonlinearity, uncertainty, and incompleteness. The production process involves numerous uncertain and abnormally fluctuating quality characteristics, making traditional control charts unsuitable for quality control in production processes due to their control principles and application conditions. Summary of the Invention

[0003] In view of this, the present invention provides a method, apparatus, medium and equipment for constructing quality control charts based on fuzzy numbers, which are used to continuously evaluate uncertain quality characteristics during the manufacturing process, thereby making accurate judgments on random errors in the manufacturing process, reflecting the actual situation of product quality more accurately, and thus being more suitable for practical use.

[0004] To achieve the first objective mentioned above, the technical solution of the quality control map construction method based on fuzzy numbers provided by this invention is as follows:

[0005] The method for constructing quality control maps based on fuzzy numbers provided by this invention includes the following steps:

[0006] Based on the key processes in the manufacturing process, the key quality characteristics corresponding to the key processes are obtained, wherein the key quality characteristics corresponding to the key processes are controlled by control charts.

[0007] Based on the key quality characteristics corresponding to the key process, define fuzzy numbers that satisfy the conditions, and determine the possible range of values ​​for the fuzzy numbers that satisfy the conditions.

[0008] By conducting quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that meet the conditions, the possible range of values ​​for the fuzzy numbers that meet the conditions, and the uncertain quality characteristics in the production process, a control chart corresponding to the uncertain quality characteristics is constructed.

[0009] The quality control chart construction method based on fuzzy numbers provided by this invention can be further implemented using the following technical measures.

[0010] Preferably, based on the key quality characteristics corresponding to the key process, a fuzzy number satisfying the conditions is defined, and determining the possible value range of the fuzzy number satisfying the conditions specifically includes the following steps:

[0011] For the key quality characteristics corresponding to the key process, obtain expert scores for the key quality characteristics corresponding to the key process, and obtain the scoring results. The scoring factors involved in the scoring results include language variables and quality levels.

[0012] Based on the scoring results, an expert scoring sample library is constructed.

[0013] Based on the expert rating sample library, a fuzzy number that meets the conditions is defined, and the possible range of values ​​for the fuzzy number that meets the conditions is determined.

[0014] Preferably, by performing quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that satisfy the conditions, the possible value range of the fuzzy numbers that satisfy the conditions, and the uncertain quality characteristics in the production process, constructing a control chart corresponding to the uncertain quality characteristics specifically includes the following steps:

[0015] Data is extracted from the expert rating sample database, wherein the data is a set of real numbers in a fuzzy set;

[0016] Based on the set of real numbers of the fuzzy set, obtain the center value of the set of real numbers, and construct the scoring range of the quality characteristics of the key quality features corresponding to the key process.

[0017] Based on the distance between the scores in the expert scoring sample library and the center value of the real number set, a distance matrix is ​​constructed. The number of rows in the distance matrix is ​​the same as the original bisection matrix in the expert scoring sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features of the key processes in the production and manufacturing process.

[0018] Based on the distance matrix, the comparison results are obtained by comparing the scores with the range one by one according to the importance of the expert scores in the set of real numbers and the average distance of the scores in the expert score sample library from the center value of the set of real numbers.

[0019] Based on the comparison results, a comparison mean matrix is ​​constructed;

[0020] Based on the comparison mean matrix and the value range of uncertain quality characteristics in the production process, a fuzzy set is constructed, and the fuzzy number in the fuzzy set is a closed interval belonging to the interval (0,1].

[0021] Based on the fuzzy number, a quality control chart based on the fuzzy number is constructed.

[0022] Preferably, in a step where a distance matrix is ​​constructed based on the distances between the scores in the expert scoring sample library and the center values ​​of the real number set, the distance matrix has the same number of rows as the original bisecting matrix in the expert scoring sample library and the same number of columns as the number of quality features in the key processes of the manufacturing process, the value range A, a of the quality feature scores is constructed. i ∈A, the relation rules are as follows:

[0023] If there exists a closed interval [m,n] on the fuzzy set μA(x), then M is the effective fuzzy number of the fuzzy set;

[0024] Suppose I and J are fuzzy numbers belonging to two fuzzy sets μI(x) and μJ(y), respectively. On the real number axes X and Y, I and J are the fuzzy numbers on those sets. Then I*J (* represents {+, -, x, / }) can define a fuzzy region Z, and their membership function is μ. i*j (z)=∨x*y=z[μ i (x)∧μ j [x], the geometric shape of this membership function is usually represented by fuzzy data such as trigonometric, trapezoidal, and normal distribution. Assuming there are four parameters a, b, c, and d, the definition formula of the membership function of their trapezoidal fuzzy numbers is as follows:

[0025]

[0026] As a preferred method, the method for constructing a trapezoidal fuzzy membership degree control chart includes the following steps:

[0027] The method for calculating the relative distance matrix is ​​as follows:

[0028] Determine the center value of set A, and determine the importance based on the distance from the center value; that is, the ratings closer to the center value have higher importance. Use a distance matrix D to represent the mutual distances, and construct the distance matrix D = [d oi ] m.n This matrix has the same number of rows as the original rating matrix and the same number of columns as the number of quality features. Therefore, for each rating value a in set A... i The distance from the center is equal to d. io =|a i -a o Calculate the average distance between them.

[0029] d i =Σ n o＝1 d io / (n-1)

[0030] di is used to measure proximity to the center. d i The smaller the value, the closer it is to the center;

[0031] The method for constructing the mean matrix is ​​as follows:

[0032] To determine the importance of each rating ai in set A, the average distance di between them is compared with ai one by one, resulting in a comparison matrix C, where C = [c oi ] m.n c oi It is a i With a o Importance comparison, using σ i Marked as a i Importance, i.e., a i The weights;

[0033] The method for calculating the fuzzy mode of the rating is as follows:

[0034] Assume σ i For a i The actual importance of , if we consider n as a quality feature of A, then σ is the corresponding quality feature vector. If σ i The equation is: If it is established, then σ i Is assigned to a i The weight of the fuzzy mode is then:

[0035] The method for calculating the left and right distribution coefficients of the fuzzy mode is as follows:

[0036] The distribution of fuzzy numbers can be determined by the distribution coefficients on the left and right sides. This is done by identifying the breakpoints on both sides and then calculating the variance of the fuzzy numbers.

[0037]

[0038] Then, substituting the previously obtained weights σj into the formula, we obtain the weighted variance of the fuzzy numbers:

[0039]

[0040] Assuming L and R represent the left and right feature numbers of the fuzzy number F, respectively, the distribution coefficient can be obtained as follows:

[0041]

[0042] The ratio of the left and right stretching of the fuzzy number is P:

[0043]

[0044] The trapezoidal membership parameters (a, b, c, d) of the quality characteristics can then be calculated as follows:

[0045] a = 2 * [k * (mL) + L] - [p * (mL) + L]

[0046] b=p*(mL)+L

[0047] c = (1-p)*(Rm) + m

[0048] d=2*[(1-k)*(Rm)+m]-[(1-p)*(Rm)+m]

[0049] Using a trapezoidal set of graphics, we obtain the geometric figure of the parameters {a,b,c,d}, where k and p are the values ​​of the upper and lower bases of the trapezoid, k represents the width of the lower base of the trapezoid, and p represents the width of the upper base of the trapezoid.

[0050] The method for constructing fuzzy number instances is as follows:

[0051] Assume that during the production process, for quality inspection items, based on the obtained expert evaluation data, a relative distance matrix is ​​obtained, as well as the mutual distance between the expert evaluation data scores;

[0052] A comparison mean matrix is ​​obtained based on the mutual distances between the scores of the expert evaluation data;

[0053] Based on the comparison mean matrix, the weight coefficients, fuzzy mode, and trapezoidal membership function values ​​are obtained, and a trapezoidal fuzzy membership degree control chart is obtained accordingly.

[0054] Preferably, the fuzzy control chart anomaly determination method includes the following steps:

[0055] First, obtain the mean information of the process of constructing the fuzzy control chart, and then use the mean of the fuzzy number samples to determine anomalies:

[0056] Suppose a fuzzy score sample of a quality characteristic is:

[0057] Take the mean of the above samples. Then the parameters in the matrix Then the cut set of the sample If the point on the control chart falls within the range of α, it is considered within the controlled range. Based on the above principle, probability and necessity measurement calculations are performed on the quality characteristics. Assuming A, B ∈ F(X), then according to the formulas for probability and necessity measurement:

[0058] Pos(B|A)=supx∈X min{μA(x),μB(x)}

[0059] Nec(B|A)=infx∈X min{μA(x),μB(x)}

[0060] In the formula, Pos and Nec are the probability measure and necessity measure of fuzzy set B under the condition that fuzzy set A occurs, respectively, and μA and μB are the membership functions of fuzzy sets A and B, respectively, based on the sample mean. available Match degree:

[0061]

[0062]

[0063] It can be known that:

[0064] If the quality characteristic measurement results meet the requirements, the probability measure cannot be less than the cutoff α (0≤α≤1), that is... The necessity measure cannot be less than the user-defined cutoff set β (0 ≤ β ≤ 0.5), that is...

[0065] Preferably, when the code is Python code, the code implementation method for uncertain quality characteristic control charts includes the following steps:

[0066] Use the skfuzzy library to create fuzzy sets, where the trimf() function is used to create a triangular fuzzy set, plot the example data y, and create a fuzzy variable named output;

[0067] The trimf() function was used to create a fuzzy output set of the fuzzy control chart;

[0068] Use the plot() function to plot the example data and the fuzzy output set, and adjust the plot title, coordinate labels, and legend.

[0069] To achieve the second objective mentioned above, the technical solution of the quality control map construction device based on fuzzy numbers provided by the present invention is as follows:

[0070] The quality control map construction device based on fuzzy numbers provided by this invention includes:

[0071] The key quality characteristic acquisition module is used to acquire the key quality characteristics corresponding to the key processes in the production and manufacturing process, wherein the key quality characteristics corresponding to the key processes are controlled by control charts.

[0072] The fuzzy number definition and value range determination module is used to define a fuzzy number that meets the conditions based on the key quality characteristics corresponding to the key process, and to determine the possible value range of the fuzzy number that meets the conditions.

[0073] The control chart construction module is used to construct a control chart corresponding to the uncertain quality characteristics by performing quality factor analysis on key processes in the production and manufacturing process, based on the fuzzy number that meets the conditions, the possible value range of the fuzzy number that meets the conditions, and the uncertain quality characteristics in the production process.

[0074] To achieve the third objective mentioned above, the technical solution of the computer-readable storage medium provided by the present invention is as follows:

[0075] The present invention provides a computer-readable storage medium storing a quality control map construction program based on fuzzy numbers. When the quality control map construction program based on fuzzy numbers is executed by a processor, it implements the steps of the quality control map construction method based on fuzzy numbers provided by the present invention.

[0076] To achieve the fourth objective mentioned above, the technical solution for the electronic device provided by this invention is as follows:

[0077] The electronic device provided by the present invention includes a memory and a processor. The memory stores a quality control map construction program based on fuzzy numbers. When the quality control map construction program based on fuzzy numbers is executed by the processor, it implements the steps of the quality control map construction method based on fuzzy numbers provided by the present invention.

[0078] This invention provides a method, apparatus, medium, and equipment for constructing quality control charts based on fuzzy numbers. These are used to continuously evaluate quality characteristics during the manufacturing process, thereby more accurately reflecting the actual product quality. Even under ideal conditions, quality data obtained during manufacturing will contain various random errors. The fuzzy number-based uncertain quality characteristic control chart modeling technology, through fuzzy set theory and mathematical statistics, uses quality control charts to judge abnormal data, solving quality control problems caused by data uncertainty, improving production efficiency and product quality, and helping enterprises achieve better quality management results. First, the uncertainty of data in uncertain quality characteristics causes traditional counting-based control chart methods to fail, making them unsuitable for directly constructing uncertain quality characteristics. The control charts constructed using fuzzy set theory provided in this invention can handle such uncertain data and provide reasonable decisions. Second, quality data in the production process is often described in linguistic form. This linguistic data can be converted into fuzzy sets using fuzzy set theory, thereby constructing control charts. Through defuzzification methods such as the weighted average method of fuzzy sets, the linguistic data can be converted into true values, thus enabling the modeling of quality control charts. Finally, the fuzzy number-based control chart for uncertain quality characteristics can determine quality control decisions based on uncertain and linguistic data, thereby improving production efficiency and product quality. This control chart can flexibly adjust control limits and inspection stringency based on different data conditions and potential anomalies during monitoring, achieving better control results. Attached Figure Description

[0079] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0080] Figure 1 This is a flowchart illustrating the steps of a quality control map construction method based on fuzzy numbers provided in an embodiment of the present invention.

[0081] Figure 2 A schematic diagram of the signal flow relationship between functional modules in the fuzzy number-based quality control map construction device provided in an embodiment of the present invention.

[0082] Figure 3 A schematic diagram of the device structure for constructing a quality control map based on fuzzy numbers for the hardware operating environment provided in this embodiment of the invention;

[0083] Figure 4 This is a schematic diagram of the fuzzy number construction process involved in the fuzzy number-based quality control map construction method provided in an embodiment of the present invention;

[0084] Figure 5 The membership function graph of trapezoidal fuzzy numbers involved in the quality control graph construction method based on fuzzy numbers provided in the embodiments of the present invention;

[0085] Figure 6 The quality control graph construction method based on fuzzy numbers provided in this embodiment of the invention is a membership function graph of trapezoidal fuzzy numbers constructed from expert scores;

[0086] Figure 7 This is a schematic diagram of the fuzzy control chart probability measure of the quality control chart construction method based on fuzzy numbers provided in the embodiments of the present invention;

[0087] Figure 8 This is a schematic diagram illustrating the necessity measure of fuzzy control charts involved in the fuzzy number-based quality control chart construction method provided in this embodiment of the invention. Detailed Implementation

[0088] To address the problems existing in the prior art, this invention provides a method, apparatus, medium, and device for constructing quality control charts based on fuzzy numbers. These are used to continuously evaluate uncertain quality characteristics during the manufacturing process, thereby accurately determining random errors in the manufacturing process and more accurately reflecting the actual product quality, making them more suitable for practical use.

[0089] Currently, count-type control charts and measure-type control charts are frequently used in production processes to analyze and control quality, thereby reflecting quality problems in the production process. The methods for constructing control charts generally involve the following steps:

[0090] 1. Determine the quality characteristics of the control: Select quality characteristics that meet the application requirements of the control chart.

[0091] 2. Data Collection: Data is accumulated to create quality control charts, which continuously reflect the quality of the analytical work. Therefore, the collected data covers changes under different conditions. Data collection generally has specific requirements regarding the amount of data and the frequency of data collection.

[0092] 3. Statistical values: When completing the data accumulation as required, the parameter values ​​of each statistical measure can be calculated according to the needs of the corresponding graph.

[0093] 4. Determine the control limits: Based on the statistical values, determine the upper and lower control limits and the center line of the control chart.

[0094] 5. Draw a quality control chart: Plot the corresponding statistical values ​​on the chart according to the measurement order, and connect the points with straight lines to form the required original quality control chart.

[0095] 6. Judgment of quality control charts: Determine whether the process is under control based on the points on the control chart.

[0096] 7. Handling Abnormal Situations: If abnormal points appear on the control chart, the process needs to be adjusted and improved.

[0097] Through arduous and persistent efforts, the inventor discovered that

[0098] The above process is generally suitable for the quality characteristics that fluctuate normally according to certain rules during the production process. The traditional count-type quality control chart can play a good role in monitoring and control.

[0099] However, modern complex equipment manufacturing is characterized by complexity, nonlinearity, uncertainty, and incompleteness. The production process involves numerous uncertain and abnormally fluctuating quality characteristics, making traditional control charts ill-suited to the quality control requirements of the production process, both in terms of control principles and application conditions. This inadequacy manifests primarily in the following ways:

[0100] 1. Traditional quality control charts are effective at monitoring quality characteristics that exhibit regular fluctuations in the production process, such as those following a normal or binomial distribution. They can construct control charts by collecting, statistically analyzing data such as the mean, median, individual values, defect rate, and number of defects, thereby monitoring the average value and range of the process and effectively controlling quality changes. However, traditional control charts are ineffective at monitoring the quality characteristics of uncertainties in the production process.

[0101] 2. Traditional control charts require a large amount of data to produce reliable graphs because they are based on the sample mean and standard deviation to calculate control limits and center lines. Insufficient data can lead to inaccurate statistical indicators in the control charts. For single-batch, small-volume, multi-variety production in industries such as aerospace, it is often impossible to provide the necessary multi-period, massive amounts of data during the manufacturing process, thus affecting control effectiveness and preventing the statistical analysis of effective values, resulting in compromised reliability and accuracy of the control charts.

[0102] 3. For anomalies in certain complex processes, such as deviations, drifts, non-periodic changes, trends, batch variations, gradual changes, cyclical and sampling differences in quality data, traditional control charts lack the ability to assess the process and therefore cannot provide sufficient control.

[0103] 4. In the manufacturing process, some processes have near-zero non-conforming quality, and the quality monitoring remains unchanged for a long time. However, traditional control charts cannot determine the trend changes in the process, which causes traditional control charts to lose their function of identifying anomalies in application.

[0104] 5. Traditional control charts assume a stable production process, with variation attributed to common causes. However, in modern complex equipment manufacturing, the continuity and reproducibility of quality parameters lead to correlations in quality data, which may contradict the assumption of process stability. Traditional control charts cannot accurately determine whether these data are abnormal or unreliable, nor can they effectively arbitrate the data. Therefore, using traditional control charts to control the continuity and reproducibility of uncertain quality characteristics is inappropriate.

[0105] Terminology Explanation

[0106] Quality characteristics: These are the characteristics of a product, process, or system that relate to requirements.

[0107] Uncertainty refers to the inability to determine the probability of complex and ever-changing things or processes occurring. In statistical processes, the situation may exist, but it cannot be completely determined.

[0108] Modeling is the process of creating a model, an abstraction of things made to understand them, and an unambiguous written description of them. The process of building a system model is also called modeling. Modeling is an important means and prerequisite for studying systems. Any process that uses a model to describe the causal relationships or interrelationships of a system falls under the category of modeling.

[0109] Control charts are charts with control limits used to analyze and determine whether each process step in a production process is in a stable state. They are an effective method for identifying and predicting abnormal fluctuations in quality.

[0110] Fuzzy sets are an extension of the classic set concept. They use membership degree to characterize the degree to which an element belongs to a set, that is, the mathematical notation μA(x) represents the degree to which element x belongs to set A.

[0111] Fuzzy number: Suppose A is a fuzzy set of the set of real numbers R, and α∈A must be a closed interval for all α∈(0,1], then α is defined as a fuzzy number.

[0112] Proximity: is a quantitative indicator that represents the degree of closeness between fuzzy sets.

[0113] Satisfaction degree: It is a quantitative indicator that represents the degree of closeness between two fuzzy sets. It focuses on the relationship between fuzzy sets and mainly refers to: inclusion, intersection, and separation.

[0114] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the fuzzy number-based quality control chart construction method, apparatus, storage medium, and device proposed according to the present invention. In the following description, different "embodiments" or "embodiments" do not necessarily refer to the same embodiment. Furthermore, features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0115] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships, such as A and / or B. Specifically, it can mean that A and B can be included at the same time, A can exist alone, or B can exist alone, and any of the above three situations can be met.

[0116] A method for constructing quality control maps based on fuzzy numbers

[0117] See appendix Figure 1 Appendix Figure 4 -Appendix Figure 8 The method for constructing a quality control map based on fuzzy numbers provided in this embodiment of the invention includes the following steps:

[0118] Step S1: Based on the key processes in the production process, obtain the key quality characteristics corresponding to the key processes. The key quality characteristics corresponding to the key processes are controlled by control charts.

[0119] Specifically, constructing control charts based on fuzzy numbers is a complex process involving numerous calculations and logical decisions. Each stage of the manufacturing process has many quality requirements, with some critical stages playing a vital role in the product's quality. Addressing the key quality characteristics of these stages primarily involves using fuzzy control charts to improve the accuracy of quality control.

[0120] (1) It is formulated by the relevant departments of the enterprise based on the characteristics of the product and related process requirements, and is one of the important indicators for evaluating product quality.

[0121] (2) Key quality characteristics must be controlled. It is necessary to develop corresponding control plans and testing methods to ensure that they meet the standard requirements, thereby guaranteeing the quality and reliability of the product.

[0122] Step S2: Based on the key quality characteristics corresponding to the key processes, define the fuzzy numbers that satisfy the conditions, and determine the possible range of values ​​for the fuzzy numbers that satisfy the conditions.

[0123] Specifically, the possible range of values ​​for fuzzy numbers is usually determined by the practical problem or the experience of experts. This embodiment of the invention uses an expert scoring sample library as the range of fuzzy numbers.

[0124] Step S3: By conducting quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that meet the conditions, the possible range of values ​​for the fuzzy numbers that meet the conditions, and the uncertain quality characteristics in the production process, construct control charts corresponding to the uncertain quality characteristics.

[0125] The quality control chart construction method based on fuzzy numbers provided by this invention is used to continuously evaluate quality characteristics during the manufacturing process, thereby more accurately reflecting the actual product quality. Even under ideal conditions, quality data obtained during manufacturing will contain various random errors. The fuzzy number-based uncertain quality characteristic control chart modeling technology, through fuzzy set theory and mathematical statistics, uses quality control charts to judge abnormal data, solving quality control problems caused by data uncertainty, improving production efficiency and product quality, and helping enterprises achieve better quality management results. First, due to the uncertainty of data in uncertain quality characteristics, traditional counting-based control chart methods fail and cannot be directly applied to the construction of uncertain quality characteristics. The control chart constructed using fuzzy set theory provided by this invention can handle such uncertain data and provide reasonable decisions. Second, quality data in the production process is often described in linguistic form. This linguistic data can be converted into fuzzy sets using fuzzy set theory, thereby constructing control charts. Through defuzzification methods such as the weighted average method of fuzzy sets, the linguistic data can be converted into true values, thus enabling the modeling of quality control charts. Finally, the fuzzy number-based control chart for uncertain quality characteristics can determine quality control decisions based on uncertain and linguistic data, thereby improving production efficiency and product quality. This control chart can flexibly adjust control limits and inspection stringency based on different data conditions and potential anomalies during monitoring, achieving better control results.

[0126] Specifically, defining fuzzy numbers that satisfy the conditions based on the key quality characteristics corresponding to the key processes, and determining the possible range of values ​​for the fuzzy numbers that satisfy the conditions, includes the following steps:

[0127] For the key quality characteristics corresponding to the key processes, expert scores are obtained for the key quality characteristics corresponding to the key processes, and the scoring results are obtained. The scoring factors involved in the scoring results include language variables and quality levels.

[0128] The process of constructing control charts corresponding to key processes in the manufacturing process, based on the fuzzy numbers that meet the conditions, the possible range of values ​​for the fuzzy numbers that meet the conditions, and the uncertain quality characteristics in the production process, specifically includes the following steps:

[0129] Data is extracted from the expert rating sample database, where the data is a set of real numbers in a fuzzy set;

[0130] Based on the set of real numbers in the fuzzy set, obtain the center value of the set of real numbers, and construct the scoring range of the quality characteristics of the key quality features corresponding to the key processes;

[0131] Based on the distance between the scores in the expert scoring sample library and the center value of the real number set, a distance matrix is ​​constructed. The number of rows in the distance matrix is ​​the same as the original bisection matrix in the expert scoring sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features of the key processes in the production and manufacturing process.

[0132] Based on the distance matrix, the comparison results are obtained by comparing the importance of expert ratings in the set of real numbers, and the average distance of the expert ratings in the expert rating sample library from the center value of the set of real numbers with the rating range one by one.

[0133] Based on the comparison results, construct a comparison mean matrix;

[0134] Based on the comparison mean matrix and the range of uncertain quality characteristics in the production process, a fuzzy set is constructed. The fuzzy number in the fuzzy set is a closed interval belonging to the interval (0,1].

[0135] Construct a quality control chart based on fuzzy numbers.

[0136] Specifically, the sample database, as the core data of fuzzy sets, provides the basic data for objective quantitative evaluation methods. It can more accurately evaluate the attributes and merits of things, improve the accuracy and reliability of evaluations, reduce human error, and enhance the accuracy of assessments of uncertain quality characteristics. The main tasks in constructing the expert scoring sample database include:

[0137] (1) First, it is necessary to determine the criteria and standards that the scoring experts must meet and the scope of the scoring.

[0138] (2) Determine the size of the sample database based on actual needs and resources, including the number of experts to be invited and the allocation of scoring tasks.

[0139] (3) Industry experts score the provided quality data according to the standards and in accordance with the prescribed scoring criteria. The scoring results are recorded in the expert scoring database.

[0140] Through the above steps, language variables, quality levels, and other data can be transformed into structured data, constructing a relatively scientific and reliable expert rating sample library, which provides important reference and basis for the subsequent construction of uncertain quality control charts.

[0141] Based on the scoring results, an expert scoring sample library was constructed.

[0142] Based on the expert scoring sample database, we define fuzzy numbers that meet the conditions, and determine the possible range of values ​​for the fuzzy numbers that meet the conditions.

[0143] In this process, a distance matrix is ​​constructed based on the distance between the scores in the expert scoring sample library and the center value of the set of real numbers. The number of rows in the distance matrix is ​​the same as the original bisection matrix in the expert scoring sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features in the key processes of the manufacturing process. This process then constructs the value range A, a of the quality feature scores. i ∈A, the relation rules are as follows:

[0144] If there exists a closed interval [m,n] on the fuzzy set μA(x), then M is the effective fuzzy number of the fuzzy set;

[0145] Suppose I and J are fuzzy numbers belonging to two fuzzy sets μI(x) and μJ(y), respectively. On the real number axes X and Y, I and J are the fuzzy numbers on those sets. Then I*J (* represents {+, -, x, / }) can define a fuzzy region Z, and their membership function is μ. i*j (z)=∨x*y=z[μ i (x)∧μ j [x], the geometric shape of this membership function is usually represented by fuzzy data such as trigonometric, trapezoidal, and normal distribution. Assuming there are four parameters a, b, c, and d, the definition formula of the membership function of their trapezoidal fuzzy numbers is as follows:

[0146]

[0147] The method for constructing a trapezoidal fuzzy membership degree control chart includes the following steps:

[0148] The method for calculating the relative distance matrix is ​​as follows:

[0149] Determine the center value of set A, and determine the importance based on the distance from the center value; that is, the ratings closer to the center value have higher importance. Use a distance matrix D to represent the mutual distances, and construct the distance matrix D = [d oi ] m.n This matrix has the same number of rows as the original rating matrix and the same number of columns as the number of quality features. Therefore, for each rating value a in set A... i The distance from the center is equal to d. io =|a i -a o Calculate the average distance between them.

[0150] d i =Σ n o＝1 d io / (n-1)

[0151] di is used to measure proximity to the center. d i The smaller the value, the closer it is to the center;

[0152] The method for constructing the comparison mean matrix is ​​as follows:

[0153] To determine the importance of each rating ai in set A, the average distance di between them is compared with ai one by one, resulting in a comparison matrix C, where C = [c oi ] m.n c oi It is a i With a o Importance comparison, using σ i Marked as a i Importance, i.e., a i The weights;

[0154] The method for calculating the fuzzy mode of the rating is as follows:

[0155] Assume σ i For a i The actual importance of , if we consider n as a quality feature of A, then σ is the corresponding quality feature vector. If σ i The equation is: If it is established, then σ i Is assigned to a i The weights are then used to determine the fuzzy mode:

[0156] The method for calculating the left and right distribution coefficients of the fuzzy mode is as follows:

[0157] The distribution of fuzzy numbers can be determined by the distribution coefficients on the left and right sides. This is done by identifying the breakpoints on both sides and then calculating the variance of the fuzzy numbers.

[0158]

[0159] Then, substituting the previously obtained weights σj into the formula, we obtain the weighted variance of the fuzzy numbers:

[0160]

[0161] Assuming L and R represent the left and right feature numbers of the fuzzy number F, respectively, the distribution coefficient can be obtained as follows:

[0162]

[0163] The ratio of the left and right stretching of the fuzzy number is P:

[0164]

[0165] The trapezoidal membership parameters (a, b, c, d) of the quality characteristics can then be calculated as follows:

[0166] a = 2 * [k * (mL) + L] - [p * (mL) + L]

[0167] b=p*(mL)+L

[0168] c = (1-p)*(Rm) + m

[0169] d=2*[(1-k)*(Rm)+m]-[(1-p)*(Rm)+m]

[0170] Using a trapezoidal set of graphics, we obtain the geometric figure of the parameters {a,b,c,d}, where k and p are the values ​​of the upper and lower bases of the trapezoid, k represents the width of the lower base of the trapezoid, and p represents the width of the upper base of the trapezoid.

[0171] The method for constructing fuzzy number instances is as follows:

[0172] Assume that during the production process, for quality inspection items, based on the obtained expert evaluation data, a relative distance matrix is ​​obtained, as well as the mutual distance between the expert evaluation data scores;

[0173] A comparison mean matrix is ​​obtained based on the mutual distances between the scores of the expert evaluation data;

[0174] Based on the comparison mean matrix, the weight coefficients, fuzzy mode, and trapezoidal membership function values ​​are obtained, and a trapezoidal fuzzy membership degree control chart is obtained accordingly.

[0175] Specifically, for example, when five experts have conducted evaluations and been included in the expert database, where G{g1=3,g2=4,g3=5,g4=5,g5=7}, the relative distance matrix can be obtained according to the formula:

[0176]

[0177] The distance between the five scores is The comparison matrix is ​​as follows:

[0178]

[0179] According to the formula, the weight coefficients σ{σ1=0.145, σ2=0.217, σ3=0.260, σ5=0.118} are known. Then, according to the formula, the fuzzy mode m=4.73, and L=0.916, R=6.889 can be calculated. Therefore, k=0.5, p=0.9. According to the formula, the trapezoidal membership function values ​​are a=1.297, b=4.349, c=4.947, d=6.674. Based on the above data, a trapezoidal fuzzy membership degree control chart is constructed.

[0180] The anomaly detection method for fuzzy control charts includes the following steps:

[0181] First, obtain the mean information of the process of constructing the fuzzy control chart, and then use the mean of the fuzzy number samples to determine anomalies:

[0182] Suppose a fuzzy score sample of a quality characteristic is:

[0183] Take the mean of the above samples. Then the parameters in the matrix Then the cut set of the sample If the point on the control chart falls within the range of α, it is considered within the controlled range. Based on the above principle, probability and necessity measurement calculations are performed on the quality characteristics. Assuming A, B ∈ F(X), then according to the formulas for probability and necessity measurement:

[0184] Pos(B|A)=supx∈X min{μA(x),μB(x)}

[0185] Nec(B|A)=infx∈X min{μA(x),μB(x)}

[0186] In the formula, Pos and Nec are the probability measure and necessity measure of fuzzy set B under the condition that fuzzy set A occurs, respectively, and μA and μB are the membership functions of fuzzy sets A and B, respectively, based on the sample mean. available Match degree:

[0187]

[0188]

[0189] It can be known that:

[0190] If the quality characteristic measurement results meet the requirements, the probability measure cannot be less than the cutoff α (0≤α≤1), that is... The necessity measure cannot be less than the user-defined cutoff set β (0 ≤ β ≤ 0.5), that is...

[0191] When the code is in Python, the code implementation method for an uncertain quality characteristic control chart includes the following steps:

[0192] Use the skfuzzy library to create fuzzy sets, where the trimf() function is used to create a triangular fuzzy set, plot the example data y, and create a fuzzy variable named output;

[0193] The trimf() function was used to create a fuzzy output set of the fuzzy control chart;

[0194] Use the plot() function to plot the example data and the fuzzy output set, and adjust the plot title, coordinate labels, and legend.

[0195] A quality control map construction device based on fuzzy numbers

[0196] The quality control map construction device based on fuzzy numbers provided by this invention includes:

[0197] The key quality characteristic acquisition module is used to acquire the key quality characteristics corresponding to the key processes in the production and manufacturing process. The key quality characteristics corresponding to the key processes are controlled by control charts.

[0198] The fuzzy number definition and value range determination module is used to define fuzzy numbers that meet the conditions based on the key quality characteristics corresponding to the key processes, and to determine the possible value range of the fuzzy numbers that meet the conditions.

[0199] The control chart construction module is used to construct control charts corresponding to uncertain quality characteristics by performing quality factor analysis on key processes in the manufacturing process, based on fuzzy numbers that meet the conditions, the possible range of values ​​for the fuzzy numbers that meet the conditions, and uncertain quality characteristics in the production process.

[0200] The fuzzy number-based quality control chart construction device provided by this invention is used to continuously evaluate quality characteristics during the manufacturing process, thereby more accurately reflecting the actual product quality. Even under ideal conditions, quality data obtained during manufacturing will contain various random errors. The fuzzy number-based uncertain quality characteristic control chart modeling technology, through fuzzy set theory and mathematical statistics, uses quality control charts to judge abnormal data, solving quality control problems caused by data uncertainty, improving production efficiency and product quality, and helping enterprises achieve better quality management results. First, due to the uncertainty of data in uncertain quality characteristics, traditional counting-based control chart methods fail and cannot be directly applied to the construction of uncertain quality characteristics. The control chart constructed using fuzzy set theory provided by this invention can handle such uncertain data and provide reasonable decisions. Second, quality data in the production process is often described in linguistic form. This linguistic data can be converted into fuzzy sets using fuzzy set theory, thereby constructing control charts. Through defuzzification methods such as the weighted average method of fuzzy sets, the linguistic data can be converted into true values, thus enabling the modeling of quality control charts. Finally, the fuzzy number-based control chart for uncertain quality characteristics can determine quality control decisions based on uncertain and linguistic data, thereby improving production efficiency and product quality. This control chart can flexibly adjust control limits and inspection stringency based on different data conditions and potential anomalies during monitoring, achieving better control results.

[0201] Computer-readable storage media

[0202] The computer-readable storage medium provided by the present invention stores a quality control map construction program based on fuzzy numbers. When the quality control map construction program based on fuzzy numbers is executed by a processor, it implements the steps of the quality control map construction method based on fuzzy numbers provided by the present invention.

[0203] The computer-readable storage medium provided by this invention is used for continuous evaluation of quality characteristics during the manufacturing process, thereby more accurately reflecting the actual product quality. Even under ideal conditions, quality data obtained during the manufacturing process contains various random errors. The fuzzy number-based uncertain quality characteristic control chart modeling technology, through fuzzy set theory and mathematical statistics, uses quality control charts to judge abnormal data, solving quality control problems caused by data uncertainty, improving production efficiency and product quality, and helping enterprises achieve better quality management results. First, due to the uncertainty of data in uncertain quality characteristics, traditional counting-based control chart methods fail and cannot be directly applied to the construction of uncertain quality characteristics. The control charts constructed using fuzzy set theory provided by this invention can handle such uncertain data and provide reasonable decisions. Second, quality data in the production process is often described in linguistic form. This linguistic data can be converted into fuzzy sets using fuzzy set theory, thereby constructing control charts. Through defuzzification methods such as the weighted average method of fuzzy sets, the linguistic data can be converted into true values, thus enabling the modeling of quality control charts. Finally, the fuzzy number-based control chart for uncertain quality characteristics can determine quality control decisions based on uncertain and linguistic data, thereby improving production efficiency and product quality. This control chart can flexibly adjust control limits and inspection stringency based on different data conditions and potential anomalies during monitoring, achieving better control results.

[0204] electronic devices

[0205] The electronic device provided by the present invention includes a memory and a processor. The memory stores a quality control map construction program based on fuzzy numbers. When the quality control map construction program based on fuzzy numbers is executed by the processor, it implements the steps of the quality control map construction method based on fuzzy numbers provided by the present invention.

[0206] This invention provides an electronic device based on fuzzy numbers for continuous evaluation of quality characteristics during the manufacturing process, thereby more accurately reflecting the actual product quality. Even under ideal conditions, quality data obtained during manufacturing contains various random errors. The fuzzy number-based uncertain quality characteristic control chart modeling technology, through fuzzy set theory and mathematical statistics, uses quality control charts to judge abnormal data, solving quality control problems caused by data uncertainty, improving production efficiency and product quality, and helping enterprises achieve better quality management results. First, the uncertainty of data in uncertain quality characteristics causes traditional counting-based control chart methods to fail, making them unsuitable for directly constructing uncertain quality characteristics. The control charts constructed using fuzzy set theory provided in this invention can handle such uncertain data and provide reasonable decisions. Second, quality data in the production process is often described in linguistic form. This linguistic data can be converted into fuzzy sets using fuzzy set theory, thereby constructing control charts. Through defuzzification methods such as the weighted average method of fuzzy sets, the linguistic data can be converted into true values, thus enabling the modeling of quality control charts. Finally, the fuzzy number-based control chart for uncertain quality characteristics can determine quality control decisions based on uncertain and linguistic data, thereby improving production efficiency and product quality. This control chart can flexibly adjust control limits and inspection stringency based on different data conditions and potential anomalies during monitoring, achieving better control results.

[0207] Reference Figure 3 , Figure 3 This is a schematic diagram of the device structure for constructing a quality control map based on fuzzy numbers in the hardware operating environment involved in the embodiments of the present invention.

[0208] like Figure 3As shown, the fuzzy number-based quality control map construction device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0209] Those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the device for constructing quality control charts based on fuzzy numbers, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0210] like Figure 3 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and a quality control map construction program based on fuzzy numbers.

[0211] exist Figure 3 In the fuzzy number-based quality control map construction device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and memory 1005 in the fuzzy number-based quality control map construction device of the present invention can be set in the fuzzy number-based quality control map construction device. The fuzzy number-based quality control map construction device calls the fuzzy number-based quality control map construction program stored in the memory 1005 through the processor 1001 and executes the fuzzy number-based quality control map construction method provided in the embodiment of the present invention.

[0212] Example

[0213] The method for constructing a quality control map based on fuzzy numbers provided in this embodiment of the invention specifically includes the following steps:

[0214] S1: Define the set of key quality characteristics.

[0215] The manufacturing process involves numerous quality requirements at each stage, with some critical stages having a vital impact on the product's quality. The key quality characteristics for these stages are selected based on their compliance with control chart application requirements.

[0216] S2: Establish an expert scoring sample database. This is to accumulate data for building quality control charts, which are used to continuously reflect the quality of analytical work.

[0217] The collected data covers changes under different conditions. Data collection generally has specific requirements regarding the amount of data and the number of data collection sessions. Let's assume we use an expert rating sample library as the range of fuzzy numbers. As the core data of fuzzy sets, the expert sample library provides the foundational data for objective quantitative evaluation methods, enabling more accurate evaluation of the attributes and merits of things, improving the accuracy and reliability of evaluations, reducing human error, and increasing the accuracy for uncertain quality characteristics.

[0218] S3: Construct a quality characteristic scoring model.

[0219] The quality characteristic matrix is ​​a crucial element in the quality control process. It is constructed by analyzing quality factors in key processes of production and manufacturing, and by defining fuzzy numbers that satisfy certain conditions based on uncertain quality characteristics. When data accumulation is completed as required, the parameter values ​​of various statistics can be calculated according to the needs of the corresponding charts. Based on the statistical values, the upper and lower control limits and center line of the control chart are determined. The corresponding statistical values ​​are then plotted on the chart in the order of measurement, and the points are connected by straight lines to form the required original quality control chart.

[0220] S4: Fuzzy control chart anomaly detection.

[0221] The process of constructing a fuzzy control chart is used to obtain the mean information, thereby enabling the identification of anomalies using the mean of the fuzzy number samples. Based on the plotted points on the control chart and the set metric values, it can be determined whether the process is under control.

[0222] S5: Code implementation of control charts for uncertain quality characteristics.

[0223] Based on the above algorithm, and utilizing a quality-related expert database and knowledge base, control charts are generated according to uncertainty and measured values. Control charts may include mean lines, upper control limits, and lower control limits.

[0224] S6; Judgment of quality control chart: Based on the control chart, determine whether the score exceeds the control range.

[0225] If the results exceed the controllable range, it indicates a quality problem requiring corresponding adjustments and improvements. Based on this, technologies such as enterprise knowledge bases and artificial intelligence can be used to establish a correspondence between relevant product quality characteristics and scoring samples, using this as training data to ultimately achieve online automatic monitoring of fuzzy number quality control charts.

[0226] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0227] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for constructing quality control maps based on fuzzy numbers, characterized in that, Includes the following steps: Based on the key processes in the manufacturing process, the key quality characteristics corresponding to the key processes are obtained, wherein the key quality characteristics corresponding to the key processes are controlled by control charts. Based on the key quality characteristics corresponding to the key process, define fuzzy numbers that satisfy the conditions, and determine the possible range of values ​​for the fuzzy numbers that satisfy the conditions. By conducting quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that meet the conditions, the possible range of values ​​for the fuzzy numbers that meet the conditions, and the uncertain quality characteristics in the production process, a control chart corresponding to the uncertain quality characteristics is constructed. Based on the key quality characteristics corresponding to the key processes, a fuzzy number that satisfies the conditions is defined, and the possible value range of the fuzzy number that satisfies the conditions is determined specifically by the following steps: For the key quality characteristics corresponding to the key process, obtain expert scores for the key quality characteristics corresponding to the key process, and obtain the scoring results. The scoring factors involved in the scoring results include language variables and quality levels. Based on the scoring results, an expert scoring sample library is constructed. Based on the expert rating sample library, define fuzzy numbers that meet the conditions, and determine the possible range of values ​​for the fuzzy numbers that meet the conditions. By conducting quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that satisfy the conditions, the possible value range of the fuzzy numbers that satisfy the conditions, and the uncertain quality characteristics in the production process, constructing a control chart corresponding to the uncertain quality characteristics specifically includes the following steps: Data is extracted from the expert rating sample database, wherein the data is a set of real numbers in a fuzzy set; Based on the set of real numbers of the fuzzy set, obtain the center value of the set of real numbers, and construct the scoring range of the quality characteristics of the key quality features corresponding to the key process. Based on the distance between the scores in the expert scoring sample library and the center value of the real number set, a distance matrix is ​​constructed. The number of rows in the distance matrix is ​​the same as the number of rows in the original scoring matrix in the expert scoring sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features of the key processes in the production and manufacturing process. Based on the distance matrix, the comparison results are obtained by comparing the scores with the range one by one according to the importance of the expert scores in the set of real numbers and the average distance of the scores in the expert score sample library from the center value of the set of real numbers. Based on the comparison results, a comparison mean matrix is ​​constructed; Based on the comparison mean matrix and the value range of uncertain quality characteristics in the production process, a fuzzy set is constructed, wherein the fuzzy number in the fuzzy set is a closed interval belonging to the interval (0, 1]. Based on the fuzzy number, a quality control chart based on the fuzzy number is constructed.

2. The method for constructing a quality control map based on fuzzy numbers according to claim 1, characterized in that, Based on the distances between the scores in the expert rating sample library and the center values ​​of the real number set, a distance matrix is ​​constructed. The number of rows in the distance matrix is ​​the same as the original bisection matrix in the expert rating sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features in the key processes of the manufacturing process. In this process, the value range A of the quality feature scores is constructed. The relationship rules are as follows: If in fuzzy set If there exists a closed interval [m,n], then M is the effective fuzzy number of the fuzzy set; Suppose that I and J belong to two fuzzy sets respectively. and Let fuzzy numbers be defined on the real number axes X and Y, with I and J representing the fuzzy numbers on those axes respectively. ( If we define a fuzzy region Z for {+, -, x, / }, then we can know their membership functions as follows: Typically, the geometric shape of this membership function is represented by fuzzy data such as trigonometric, trapezoidal, and normal distributions. Assuming there are four parameters a, b, c, and d, the definition formula for the membership function of their trapezoidal fuzzy numbers is as follows: .

3. The method for constructing a quality control map based on fuzzy numbers according to claim 1, characterized in that, When the code is in Python, the code implementation method for an uncertain quality characteristic control chart includes the following steps: Use the skfuzzy library to create fuzzy sets, where the trimf() function is used to create a triangular fuzzy set, plot the example data y, and create a fuzzy variable named output; The trimf() function was used to create a fuzzy output set of the fuzzy control chart; Use the plot() function to plot the example data and the fuzzy output set, and adjust the plot title, coordinate labels, and legend.

4. A quality control map construction device based on fuzzy numbers, characterized in that, include: The key quality characteristic acquisition module is used to acquire the key quality characteristics corresponding to the key processes in the production and manufacturing process, wherein the key quality characteristics corresponding to the key processes are controlled by control charts. The fuzzy number definition and value range determination module is used to define a fuzzy number that meets the conditions based on the key quality characteristics corresponding to the key process, and to determine the possible value range of the fuzzy number that meets the conditions. The control chart construction module is used to construct a control chart corresponding to the uncertain quality characteristics by performing quality factor analysis on key processes in the production and manufacturing process, based on the fuzzy number that meets the conditions, the possible value range of the fuzzy number that meets the conditions, and the uncertain quality characteristics in the production process. Based on the key quality characteristics corresponding to the key processes, a fuzzy number that satisfies the conditions is defined, and the possible value range of the fuzzy number that satisfies the conditions is determined specifically by the following steps: For the key quality characteristics corresponding to the key process, obtain expert scores for the key quality characteristics corresponding to the key process, and obtain the scoring results. The scoring factors involved in the scoring results include language variables and quality levels. Based on the scoring results, an expert scoring sample library is constructed. Based on the expert rating sample library, define fuzzy numbers that meet the conditions, and determine the possible range of values ​​for the fuzzy numbers that meet the conditions. By conducting quality factor analysis on key processes in the manufacturing process, and based on the fuzzy numbers that satisfy the conditions, the possible value range of the fuzzy numbers that satisfy the conditions, and the uncertain quality characteristics in the production process, constructing a control chart corresponding to the uncertain quality characteristics specifically includes the following steps: Data is extracted from the expert rating sample database, wherein the data is a set of real numbers in a fuzzy set; Based on the set of real numbers of the fuzzy set, obtain the center value of the set of real numbers, and construct the scoring range of the quality characteristics of the key quality features corresponding to the key process. Based on the distance between the scores in the expert scoring sample library and the center value of the real number set, a distance matrix is ​​constructed. The number of rows in the distance matrix is ​​the same as the number of rows in the original scoring matrix in the expert scoring sample library, and the number of columns in the distance matrix is ​​the same as the number of quality features of the key processes in the production and manufacturing process. Based on the distance matrix, the comparison results are obtained by comparing the scores with the range one by one according to the importance of the expert scores in the set of real numbers and the average distance of the scores in the expert score sample library from the center value of the set of real numbers. Based on the comparison results, a comparison mean matrix is ​​constructed; Based on the comparison mean matrix and the value range of uncertain quality characteristics in the production process, a fuzzy set is constructed, wherein the fuzzy number in the fuzzy set is a closed interval belonging to the interval (0, 1]. Based on the fuzzy number, a quality control chart based on the fuzzy number is constructed.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a quality control map construction program based on fuzzy numbers, which, when executed by a processor, implements the steps of the quality control map construction method based on fuzzy numbers as described in any one of claims 1-3.

6. An electronic device, characterized in that, The device includes a memory and a processor. The memory stores a fuzzy number-based quality control map construction program. When executed by the processor, the fuzzy number-based quality control map construction program implements the steps of the fuzzy number-based quality control map construction method according to any one of claims 1-3.

Images