A product quality assessment method and system based on big data

Abstract

The present invention relates to the field of product evaluation, and more specifically, to a product quality evaluation method and system based on big data. The method comprises: collecting multi-dimensional detection data, obtaining a quality evaluation value of a product, and calculating the degree of abnormality; calculating the degree of influence of the detection data of a target dimension on a target product; respectively calculating the correlation between the detection data of the target dimension and the detection data of each dimension, and using the global influence degree to correct the correlation coefficient correction value, constructing a correlation coefficient correction value matrix, and using matrix decomposition to obtain the weight of the detection data of the target dimension during quality evaluation; respectively multiplying the weight by the detection data of the corresponding dimension to obtain new multi-dimensional detection data to complete product quality evaluation. Through the technical solution of the present invention, the accuracy and efficiency of product quality evaluation can be improved.

Description

Technical Field

[0001] The present invention relates to the field of product evaluation, and more specifically, to a product quality evaluation method and system based on big data. Background Art

[0002] In today's highly competitive market, product quality is a key factor for companies to gain consumer trust, establish brand reputation, and maintain market competitiveness. High-quality products not only meet consumer needs but also reduce after-sales service costs and enhance customer satisfaction and loyalty. Traditional product quality assessment methods typically rely on limited sample testing, manual inspection, and empirical judgment, which have many limitations. The development of big data technology has enabled the collection and analysis of vast amounts of product quality-related data, providing new solutions for product quality assessment. Through big data analysis, product quality can be assessed more comprehensively and objectively, potential problems can be identified promptly, and production processes can be optimized.

[0003] The existing Chinese patent application document with publication number CN115293586A discloses a product quality assessment method and device, which includes: classifying the products to be evaluated, including physical products and non-physical products; determining the quality assessment indicators of the two types of products respectively, and obtaining data for each quality assessment indicator; determining the weight of each quality assessment indicator; performing a comprehensive weight calculation based on the indicator weight value; and analyzing the assessment results based on the obtained weight calculation results.

[0004] However, during the product manufacturing process, there are situations where even if all other factors meet the standards, certain key defects will cause the product's quality assessment results to decline, and the product will be assessed as unqualified. Such defects may be uncommon and occur infrequently. At this time, directly determining the product evaluation index data will affect the accuracy of the quality assessment results. Summary of the Invention

[0005] In order to solve the problem of low accuracy of quality assessment results, the present invention proposes a product quality assessment method and system based on big data.

[0006] In a first aspect, the present invention discloses a product quality assessment method based on big data, comprising: collecting multi-dimensional detection data, obtaining a quality assessment value of the product, taking any dimension as a target dimension, taking any product as a target product, and calculating the degree of abnormality; mapping the quality assessment value through negative correlation, and multiplying the product with the degree of abnormality as the degree of influence of the detection data of the target dimension on the target product, traversing to obtain the degree of influence of the detection data of the target dimension on each product, and taking the maximum value of the degree of influence as the global degree of influence of the detection data of the target dimension; respectively calculating the correlation between the detection data of the target dimension and the detection data of each dimension, and using the global degree of influence to correct them to obtain a correlation coefficient correction value, constructing a correlation coefficient correction value matrix, and using matrix decomposition to obtain the weight of the detection data of the target dimension in quality assessment; respectively multiplying the weights by the detection data of the corresponding dimensions to obtain new multi-dimensional detection data to complete product quality assessment.

[0007] By comprehensively considering multi-dimensional test data and product quality assessments, key influencing factors and potential anomalies are effectively identified. Furthermore, through negative correlation mapping and calculation of global impact, the specific impact of each test data point on product quality is precisely quantified. By constructing a correlation coefficient correction value matrix and performing matrix decomposition, the weights of each dimension of data are intelligently adjusted, improving the accuracy and reliability of the assessment.

[0008] Preferably, the abnormality degree satisfies the relationship:

[0009] , Indicates product In dimension The abnormality of the detection data, Indicates product In dimension The detection value of the detection data, Indicates that all products in dimension The mean of the detection values ​​of the detection data.

[0010] By comparing the test value of a single product with the mean test value of all products in that dimension, we can determine whether the performance of each product in that dimension deviates from the average level, and highlight those products that perform abnormally in quality indicators.

[0011] Preferably, the degree of abnormality also includes: calculating the first variance of the detection data of all products in the target dimension, calculating the second variance of the detection data of all products except the target product in the target dimension; calculating the ratio of the second variance to the first variance, and taking the absolute difference between the ratio and 1 as the degree of abnormality.

[0012] Being able to comprehensively evaluate whether the product's performance in the target dimension deviates from the normal range helps to improve the stability and reliability of product quality, and can also detect and solve problems in a timely manner during the production process.

[0013] Preferably, the separately calculating the correlation between the detection data of the target dimension and the detection data of each dimension includes: constructing the detection data of all products in the same dimension into a sequence; and using the Pearson correlation coefficient to separately calculate the correlation between the target dimension and each dimension.

[0014] Being able to quantify the strength of the linear relationship between the target dimension and other dimensions helps to identify which dimensions have a stronger correlation with the target dimension, so that more attention can be paid to these dimensions in product quality assessment.

[0015] Preferably, the correlation coefficient correction value satisfies the relationship:

[0016] , Representation Dimension and dimensions The correlation coefficient correction value of the detection data, Representation Dimension and dimensions The correlation coefficient of the test data, Representation Dimension The global impact of the detection data, Representation Dimension The global impact of the detection data, Indicates the total number of dimensions, and Can be equal.

[0017] Preferably, the weight satisfies the relationship:

[0018] , Representation Dimension The weight of Representation Dimension The eigenvalues ​​of Indicates the total number of dimensions.

[0019] Preferably, completing the product quality assessment includes: using new multi-dimensional detection data to obtain an updated quality assessment value of the product; when the updated quality assessment value is less than a preset threshold, generating and sending an alarm signal.

[0020] In a second aspect, the present invention discloses a product quality assessment system based on big data, comprising: a processor; and a memory, wherein the memory stores computer instructions. When the computer instructions are executed by the processor, the system executes the above-mentioned product quality assessment method based on big data.

[0021] Beneficial effects of the present invention:

[0022] The degree of anomaly for each product in each test dimension is calculated. Then, by combining negative correlation mapping with the degree of anomaly, the global importance of each test data on product quality is determined. The Pearson correlation coefficient is used to quantify the correlations between dimensions, and the correlation coefficients are corrected based on the degree of global impact, constructing a more accurate correlation coefficient correction value matrix. Matrix decomposition techniques are used to obtain weights that reflect the relative importance of each dimension in the overall quality assessment. Combining these corrected weights with the corresponding test data yields new multi-dimensional test data, which can improve the accuracy and efficiency of product quality assessments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0024] Figure 1 This is a flowchart of a product quality assessment method based on big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0026] It should be understood that when the terms "first," "second," and the like are used in the claims, description, and drawings of the present invention, they are merely used to distinguish between different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the description and claims of the present invention indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0027] The present invention provides a product quality assessment method based on big data. Figure 1 As shown, a product quality assessment method based on big data includes steps S1 to S3, which are described in detail below.

[0028] S1, collects multi-dimensional detection data, obtains product quality assessment values, and calculates the degree of abnormality.

[0029] In one embodiment, collecting multi-dimensional test data is crucial during the product quality assessment process. This test data includes not only traditional structural strength and durability testing, but also covers aspects such as appearance and structure inspection, electromagnetic compatibility testing, battery life, signal reception quality, and sound and image clarity. For example, structural strength testing can ensure that the product design will not break down under the physical stress of daily use.

[0030] Appearance and structural inspections ensure that products meet visual and tactile quality standards. This includes checking for surface defects, color uniformity, and ergonomic design. Electromagnetic compatibility testing assesses the stability of electronic products under electromagnetic interference, which is particularly important for modern electronic devices, as they are often used in complex electromagnetic environments.

[0031] Battery life testing is particularly critical for portable electronic devices, as it directly impacts product usability. Signal reception quality, sound, and image clarity testing are directly related to user experience, particularly in multimedia devices, where these metrics directly impact product performance.

[0032] By using existing technologies, such as data analysis and machine learning algorithms, valuable information can be extracted from these multi-dimensional inspection data to obtain product quality assessment values.

[0033] Calculate the degree of abnormality, which satisfies the relationship:

[0034] , Indicates product In dimension The abnormality of the detection data, Indicates product In dimension The detection value of the detection data, Indicates that all products in dimension The mean of the detection values ​​of the detection data.

[0035] In another embodiment, the degree of abnormality also includes: calculating the first variance of the detection data of all products in the target dimension, calculating the second variance of the detection data of all products except the target product in the target dimension; calculating the ratio of the second variance to the first variance, and taking the absolute difference between the ratio and 1 as the degree of abnormality.

[0036] S2, after the quality assessment value is mapped through negative correlation, the product of the quality assessment value and the degree of abnormality is used as the degree of influence of the target dimension detection data on the target product. The degree of influence of the target dimension detection data on each product is traversed to obtain the maximum value of the influence degree as the global influence degree of the target dimension detection data.

[0037] It should be noted that when the quality of a product is evaluated through detection data from various dimensions, since the frequency of occurrence of detection data from some dimensions is low, the quality evaluation value will also decrease once the detection data of that dimension is abnormal. This is because the detection data of that dimension has a large impact on the quality evaluation value and should have a greater weight when evaluating the quality of the product.

[0038] For the test data of the same dimension, selecting the maximum impact degree as the global impact degree can select the test value corresponding to the product that is most sensitive to the test data of this dimension, thereby increasing the weight of the test data of this dimension when evaluating the quality of the product.

[0039] S3, respectively calculates the correlation between the detection data of the target dimension and the detection data of each dimension, and uses the global influence degree to correct it to obtain the correlation coefficient correction value, constructs the correlation coefficient correction value matrix, and uses matrix decomposition to obtain the weight of the detection data of the target dimension in quality assessment.

[0040] It should be noted that when calculating the quality assessment value of the product, there will be influence between the detection data of each dimension. Therefore, the correlation coefficient between the detection data of each dimension is calculated, and then the global influence degree of the detection data of a single dimension is used to correct the correlation coefficient between the detection data of the two dimensions to obtain the correlation coefficient correction value.

[0041] During the production process, the interaction between various test data dimensions has a crucial impact on final product quality. Due to the correlation between these data dimensions, an abnormality in one factor can trigger a chain reaction, affecting other factors and ultimately affecting the product's quality test results. For example, if the material's hardness doesn't meet the standard, it may cause cracks in the product during processing.

[0042] In one embodiment, the inspection data of all products in the same dimension are constructed into a sequence; and the correlation between the target dimension and each dimension is calculated using the Pearson correlation coefficient.

[0043] The correlation coefficient correction value satisfies the relationship:

[0044] , Representation Dimension and dimensions The correlation coefficient correction value of the detection data, Representation Dimension and dimensions The correlation coefficient of the test data, Representation Dimension The global impact of the detection data, Representation Dimension The global impact of the detection data, Indicates the total number of dimensions, and Can be equal.

[0045] The minimum value is chosen because the correlation coefficient between the test data of two dimensions represents the mutual influence relationship between the test data of the two dimensions, and is not greatly affected by the global influence value of the test data of a single dimension. The purpose of correcting the correlation coefficient with the global influence value is to adjust the weight of the test data of a single dimension when evaluating product quality.

[0046] The greater the global impact, the more important the detection data of the corresponding dimension is. When calculating the correlation between the detection data of two dimensions, the lower limit of the global impact of the detection data of the two dimensions should be used as the correction value of the correlation coefficient of the detection data of the two dimensions, because the lower limit determines the weakest impact in the detection data of the two dimensions.

[0047] A correlation coefficient correction value matrix is ​​constructed, and matrix decomposition is used to obtain the eigenvalues ​​and eigenvectors of the detection data of the target dimension. The size of the eigenvalue can indicate the importance of the detection data of the relevant dimension in product quality assessment.

[0048] The weights used in quality assessment are calculated based on the detection data of the eigenvalue target dimension, and the weights satisfy the relationship:

[0049] , Representation Dimension The weight of Representation Dimension The eigenvalues ​​of Indicates the total number of dimensions.

[0050] S4, multiply the weights by the test data of the corresponding dimensions respectively to obtain new multi-dimensional test data to complete the product quality assessment.

[0051] In one embodiment, the updated quality assessment value of the product is obtained using the new multi-dimensional detection data; when the updated quality assessment value is less than a preset threshold, an alarm signal is generated and sent. Exemplarily, the preset threshold is set to 0.6.

[0052] An embodiment of the present invention also discloses a product quality assessment system based on big data, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a product quality assessment method based on big data according to the present invention is implemented.

[0053] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0054] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connected to the device.

[0055] While this specification has shown and described several embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

[0056] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A product quality assessment method based on big data, characterized in that: include: Collect multi-dimensional test data to obtain product quality assessment values; take any dimension as the target dimension and any product as the target product, calculate the degree of abnormality, and satisfy the relationship: , Indicates product In dimension The abnormality of the detection data, Indicates product In dimension The detection value of the detection data, Indicates that all products in dimension The mean value of the detection data; After the quality assessment value is mapped through negative correlation, the product of the quality assessment value and the degree of abnormality is used as the impact of the target dimension's detection data on the target product. The impact of the target dimension's detection data on each product is obtained by traversing, and the maximum impact is used as the global impact of the target dimension's detection data. Calculate the correlation between the detection data of the target dimension and the detection data of each dimension respectively; The correlation coefficient correction value is obtained by using the global influence degree to satisfy the relationship: , Representation Dimension and dimensions The correlation coefficient correction value of the detection data, Representation Dimension and dimensions The correlation coefficient of the test data, Representation Dimension The global impact of the detection data, Representation Dimension The global impact of the detection data, Indicates the total number of dimensions, and can be equal; construct a correlation coefficient correction value matrix, and use matrix decomposition to obtain the weight of the target dimension detection data in quality assessment; Multiply the weights by the test data of the corresponding dimensions to obtain new multi-dimensional test data to complete the product quality assessment.

2. A product quality assessment method based on big data according to claim 1, characterized in that: The degree of abnormality also includes: Calculate the first variance of the test data of all products in the target dimension, and calculate the second variance of the test data of all products except the target product in the target dimension; Calculate the ratio of the second variance to the first variance, and use the absolute difference between the ratio and 1 as the degree of abnormality.

3. The product quality assessment method based on big data according to claim 1, characterized in that: The respectively calculating the correlation between the detection data of the target dimension and the detection data of each dimension includes: Construct the inspection data of all products in the same dimension into sequences; The Pearson correlation coefficient was used to calculate the correlation between the target dimension and each dimension.

4. The product quality assessment method based on big data according to claim 1, characterized in that: The weights satisfy the relationship: , Representation Dimension The weight of Representation Dimension The eigenvalues ​​of Indicates the total number of dimensions.

5. The product quality assessment method based on big data according to claim 1, characterized in that: The product quality assessment includes: Use new multi-dimensional test data to obtain updated product quality assessment values; When the updated quality assessment value is less than the preset threshold, an alarm signal is generated and sent.

6. A product quality assessment system based on big data, characterized in that: include: processor; and A memory storing computer instructions, wherein when the computer instructions are executed by a processor, the system executes a product quality assessment method based on big data according to any one of claims 1 to 5.

Images