Industrial data quality quantitative evaluation method and device

Abstract

The invention provides an industrial data quality quantitative evaluation method and device. The problem that data quality cannot be scientifically measured, compared and improved due to scattered evaluation dimensions, non-uniform index quantification and subjective weight in the prior art is solved. The industrial data quality quantitative evaluation method comprises the following steps: acquiring multi-dimensional data from an industrial data source; quantizing the multi-dimensional data according to a plurality of preset quality dimensions, and determining a quantized value corresponding to each quality dimension; and according to preset weight configuration, performing weighted aggregation on the quantized values of the plurality of quality dimensions to generate a comprehensive score. Uniform quality measurement is realized through a standardized evaluation system, the problem of data comparability is solved through standardized mapping, result objectivity is guaranteed through scientific right confirmation, a treatment closed loop is formed in combination with short-board diagnosis, and effective support is provided for industrial enterprise data asset management.

Description

Technical Field

[0001] This application relates to the field of electronic data processing technology, and in particular to a method and apparatus for quantitative evaluation of industrial data quality. Background Technology

[0002] With the acceleration of industrial digitalization, data has become a core element driving production and decision-making. Driven by policies promoting the Industrial Internet, intelligent manufacturing, and data-driven development, data is gradually becoming a crucial link connecting equipment, processes, and decision-making, and a vital foundation for measuring the level of intelligence in industrial enterprises. However, the governance and evaluation of industrial data quality still face significant challenges. First, existing evaluation methods are mostly qualitative assessments or rely on scattered indicators, lacking a unified and scalable quantitative system. This makes it impossible to compare data quality across different enterprises and industries, hindering the market circulation of data elements. The lack of unified quality evaluation standards makes it difficult for enterprises to measure data quality and achieve cross-industry and cross-system quality benchmarking and value assessment. Second, due to the diverse dimensions and complex characteristics of industrial data indicators (such as the coexistence of positive, negative, and interval indicators), the lack of standardized mapping methods makes it difficult to fairly quantify and aggregate multiple indicators within the same system, leading to distorted evaluation results. Third, the determination of weights in the evaluation process is highly subjective, lacking constraints from objective mathematical methods such as consistency checks, resulting in insufficient scientific validity and credibility of the evaluation results. Meanwhile, existing evaluation results lack interpretability, making it difficult for enterprises to identify specific quality shortcomings from the overall score and translate them into effective governance actions. Furthermore, existing technologies lack versatility across multiple industrial scenarios, hindering rapid migration and deployment in differentiated environments such as steel, chemicals, and energy, thus limiting large-scale application.

[0003] With the advancement of national policies on data elementization, industrial enterprises have an urgent need for reliable, quantifiable, and operable evaluation of data quality. Therefore, there is a pressing need in this field for a systematic, standardized, quantifiable, and cross-industry-capable technical solution for industrial data quality evaluation. Summary of the Invention

[0004] This application proposes a method and apparatus for quantitative evaluation of industrial data quality, which solves the problems of existing technologies where data quality cannot be scientifically measured, compared and improved due to scattered evaluation dimensions, inconsistent index quantification and subjective weighting.

[0005] In a first aspect, embodiments of this application provide a method for quantitative evaluation of industrial data quality, including the following steps: Acquire multidimensional data from industrial data sources; The multidimensional data is quantized according to multiple preset quality dimensions to determine the quantized value for each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values. Based on the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

[0006] In one embodiment, the quantization process specifically includes: Identify the data orientation of sub-indicators corresponding to the quality dimension in the multidimensional data; the data orientation includes positive indicators, negative indicators, or interval indicators; Based on the identified data orientation, a preset target mapping function is determined; The score value of the sub-index is calculated based on the target mapping function.

[0007] Furthermore, the target mapping function includes: The higher the value, the higher the score of the first-class mapping function; The second type of mapping function awards higher scores for smaller values. The third type of mapping function is one in which the score is highest when the value is within the target range and decreases when it deviates from the range.

[0008] In one embodiment, determining the weight configuration includes a consistency check; The weight configuration is effective when the consistency ratio (CR) is less than a set threshold.

[0009] In one embodiment, before the weighted aggregation of quantized values, the following steps are included: The quantified value is corrected based on the confidence factor of the data source and / or the data sample size.

[0010] In one embodiment, the step of: The quality level is determined by comparing the score with a preset level threshold.

[0011] In one embodiment, the step of: Compare the quantified values ​​of multiple quality dimensions with preset quality thresholds; In response to a quantification value being less than the corresponding quality threshold, a data governance recommendation corresponding to the quantification value is determined based on a preset rule base.

[0012] Secondly, embodiments of this application also provide an industrial data quality quantification evaluation device for implementing the industrial data quality quantification evaluation method described in any embodiment of the first aspect, comprising: an acquisition module for acquiring multidimensional data from an industrial data source; a determination module for quantifying the multidimensional data according to multiple preset quality dimensions to determine a quantized value corresponding to each quality dimension; the quantification process includes uniformly mapping raw data with different dimensions to dimensionless score values; and a calculation module for weighted aggregation of the quantized values ​​of multiple quality dimensions according to a preset weight configuration to generate a comprehensive score.

[0013] Thirdly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any one of the embodiments of the first aspect.

[0014] Fourthly, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any embodiment of the first aspect.

[0015] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: This application constructs a unified "five-dimensional" evaluation system and employs a standardized positive / negative index quantification mapping method to transform raw data with different dimensions and characteristics into dimensionless standardized scores. This fundamentally solves the industry problem of the inability to make horizontal comparisons of industrial data quality due to inconsistent evaluation standards, providing a reliable quality benchmark for the circulation and trading of data elements. By introducing the Analytic Hierarchy Process (AHP) and combining it with a rigorous consistency check mechanism to determine weights, the application effectively overcomes the shortcomings of excessive subjectivity in weight allocation in traditional methods. Simultaneously, robustness strategies such as confidence level correction and sample size penalties significantly enhance the objectivity, robustness, and industry credibility of the evaluation results. The solution not only outputs a single comprehensive score but also provides multi-dimensional radar chart analysis, clear quality levels, and a specific list of quality shortcomings. More importantly, it can automatically generate targeted data governance suggestions based on a rule base, directly transforming evaluation conclusions into specific and actionable governance actions, forming a closed loop of "evaluation-diagnosis-governance," greatly improving the efficiency and effectiveness of data quality management. It possesses excellent industry adaptability and engineering application value: By pre-configured industry indicator template libraries and dynamic target values, this solution can quickly adapt to significantly differentiated industrial scenarios such as steel, mining, chemical, and power industries while maintaining the core "five-dimensional" framework. Its modular system design and device implementation support flexible deployment on local, cloud, or edge devices, providing a clear technical path and tool support for large-scale, replicable engineering applications. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart of an industrial data quality quantitative evaluation method provided in this application embodiment; Figure 2 A schematic diagram of the industrial data quality quantitative evaluation method provided in the embodiments of this application; Figure 3 A flowchart illustrating a method for determining the quality level of industrial data quality through quantitative evaluation, as provided in this application embodiment; Figure 4 A flowchart illustrating the feedback governance recommendations for a quantitative evaluation method of industrial data quality provided in this application embodiment; Figure 5 This is a structural diagram of an industrial data quality quantification evaluation device according to an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0019] To address the problems in current industrial data quality evaluation, such as fragmented evaluation dimensions, inconsistent indicator quantification, lack of objectivity in weight determination, and difficulty in interpreting and generalizing evaluation results, this invention proposes a method, apparatus, equipment, and medium for a five-dimensional quantitative evaluation of industrial data quality. This method is based on five core dimensions: completeness, accuracy, consistency, timeliness, and usability. It constructs a full-link evaluation system from indicator quantification to comprehensive scoring and grade assessment, and introduces the Analytic Hierarchy Process (AHP) and a consistency check mechanism (CR < 0.1) to ensure the scientific and verifiable nature of weight allocation. Through this system, industrial data quality can be standardized, quantified, and compared across industries, supporting data asset management and intelligent applications, and providing a reliable foundation for data-driven decision-making and industrial intelligent upgrading for enterprises.

[0020] Figure 1 A flowchart of a method for quantitative evaluation of industrial data quality provided in this application includes steps 110-130.

[0021] Step 110: Obtain multidimensional data from industrial data sources.

[0022] The multidimensional data refers to a data set composed of multiple different attribute fields obtained from industrial data sources. This "multidimensionality" emphasizes the diversity and richness of the data itself. This definition clarifies the basic form of the evaluation object: the evaluation is not aimed at a single, isolated data point, but rather at a collection of data records composed of multiple attributes that can comprehensively describe industrial entities or events.

[0023] For example, a data record from a production line may contain multiple fields such as "equipment ID", "timestamp", "temperature sensor reading", "pressure sensor reading", "vibration frequency", "equipment status code (0 normal / 1 fault)", and "production batch number". These fields together constitute "multidimensional data" describing the production event.

[0024] In this application, the term "industrial data source" broadly refers to any information system, device, sensor, or platform that generates, collects, stores, or processes data in an industrial manufacturing and production operation environment. These data sources are the original data input providers for industrial data quality assessment methods, and their diversity, heterogeneity, and real-time nature are one of the fundamental reasons for the complexity of data quality.

[0025] With the advancement of full-process digitalization in industrial enterprises' production operations, data has transformed from a "byproduct" supporting decision-making into a core production factor driving production, optimizing processes, and incubating new businesses. Its quality not only affects the stability and safety of the production process but also directly impacts the depth of process optimization, the training effectiveness of intelligent models, the efficiency of cross-departmental collaboration, and the reliability of management decisions.

[0026] The aforementioned industrial data sources collectively constitute the industrial data ecosystem, but they differ significantly in the frequency, granularity, structure, and purpose of data generation. This difference is the root cause of the complexity of data quality issues and dictates that subsequent quantitative evaluations must be adapted to the characteristics of data from different sources.

[0027] For example, motor speed data acquired from PLC sensors is high-frequency, raw, real-time data. Its quality is primarily assessed based on timeliness (delay) and accuracy (whether the value jumps). This type of data forms the basis for predictive maintenance and real-time control of equipment, and its quality defects can directly lead to production safety risks.

[0028] The number of completed work orders obtained from MES is event-driven, aggregated business data. Its quality focuses more on completeness (whether records are missing) and consistency (whether it matches the plan in ERP). This type of data is the basis for production schedule management and efficiency accounting, and its quality directly affects the accuracy of operational decisions.

[0029] Product composition analysis results obtained from LIMS are key quality attribute data, with accuracy and usability (whether the format can be parsed by downstream systems) being the most crucial quality dimensions. This type of data is directly related to product quality assessment and process optimization; its quality is the lifeline of a company.

[0030] Energy consumption monitoring data acquired from SCADA systems is periodically sampled process control data. Its quality focuses on consistency (whether the metering standards of multiple devices are consistent) and completeness (whether there are any missing points or interruptions within the sampling period). This type of data is the foundation for energy efficiency analysis and carbon emission accounting; deviations in its quality will lead to distortions in energy consumption assessments and energy-saving control strategies.

[0031] Therefore, the multidimensional data evaluated in this application is aggregated from these heterogeneous and complex data sources. However, current industrial data quality governance practices generally lack unified and scalable evaluation dimensions and indicator systems, making it impossible to compare data quality across different industries and enterprises, and hindering the flow and trading of data elements within regions or even nationwide. Faced with the ongoing national "Data Elements ×" Three-Year Action Plan, the Industrial Internet Innovation and Development Action, and the policy of industrial data assetization, industrial enterprises' demand for high-quality data is experiencing explosive growth. The subsequent quantitative evaluation steps in this application are precisely to address these challenges by systematically and standardizedly measuring the quality of multidimensional data obtained from heterogeneous data sources. Only by establishing a unified evaluation benchmark can a reliable data quality foundation be provided for the circulation of data elements, the entry of data assets into tables, and the intelligent upgrading of industry.

[0032] Step 120: Quantize the multidimensional data according to multiple preset quality dimensions to determine the quantization value corresponding to each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values.

[0033] The preset multiple quality dimensions refer to a set of evaluation angles predefined for comprehensively assessing data quality.

[0034] For example, the five-dimensional evaluation system proposed in this application includes completeness, accuracy, consistency, timeliness, and usability. These five dimensions constitute the basic framework for systematic evaluation and can comprehensively cover the key characteristics of industrial data quality.

[0035] This application proposes a five-dimensional (hereinafter referred to as "five-dimensional") evaluation system encompassing "completeness, accuracy, consistency, timeliness, and usability," and a methodology for constructing the chain of "indicator quantification → dimension aggregation → comprehensive scoring → level assessment," such as... Figure 2 As shown, a weight determination and consistency verification mechanism (AHP+CR<0.1) is given, forming an evaluation device and system that can be directly deployed in engineering.

[0036] For example, the set of dimensions described in this application:

[0037] The quantitative processing refers to the entire process of transforming raw data into standardized scores, the core of which is to solve the problem that different indicators cannot be directly compared due to their different dimensions and distributions.

[0038] This step is the key innovation of this application. By establishing a standardized quantitative method, it solves the problem in the prior art that the evaluation results cannot be compared horizontally due to inconsistent indicator standards.

[0039] In one embodiment, step 120 specifically includes: Step 120-1: Identify the data orientation of the sub-indicators corresponding to the quality dimension in the multidimensional data; the data orientation includes positive indicators, negative indicators, or interval indicators.

[0040] The sub-indices within a dimension: the j-th dimension contains n j Sub-indicators, weights ,and w ij is the weight of the i-th sub-index in the j-th dimension.

[0041] This identification process forms the basis for selecting an appropriate mapping function, ensuring the scientific rigor and relevance of the quantification process. For example, data coverage is a positive indicator; a higher value indicates better quality. Data missing rate is a negative indicator; a lower value indicates better quality. Equipment temperature values ​​may be range-based indicators, requiring a specific temperature range to obtain the highest score.

[0042] Step 120-2: Determine the preset target mapping function based on the identified data orientation.

[0043] By establishing the correspondence between indicator characteristics and mapping functions, a specialized quantification scheme is provided for different types of indicators, ensuring the rationality and comparability of the scoring results.

[0044] Furthermore, the target mapping function includes: To ensure cross-industry uniformity to the 0–100 dimension, a mapping family combining directional and distributional characteristics is adopted.

[0045] The first type of mapping function is the one where a larger value results in a higher score. This first type of mapping function is suitable for positive indicators such as data coverage and data timeliness. It uses linear piecewise saturation and other mapping methods to suppress the impact of extreme noise.

[0046] Positive indicators (the higher the value, the better) Linear piecewise saturation: If the actual value ,but S represents the indicator score (the same applies below). T represents the target threshold; for example, if the accuracy rate is >95%, then the target threshold is 95%.

[0047] For example, when the actual value of an indicator reaches or exceeds a preset target threshold T, the indicator score S can directly obtain full marks. For instance, if the target threshold T for data accuracy is set to 95%, then when the actual accuracy is greater than or equal to 95%, the indicator score S is full marks.

[0048] For example, to suppress extreme noise, an upper limit of 100 and a lower limit of 0 can be set, and a soft thresholding function can be selected. To enhance robustness. This is a soft threshold that fluctuates around the standard value of T.

[0049] The second type of mapping function is the one where the smaller the value, the higher the score. This second type of mapping function is suitable for inverse indicators such as data missing rate and outlier percentage. It uses mapping methods such as inverse standardization or logarithmic transformation to handle heavy-tailed distributions.

[0050] Inverse metrics (the smaller the value, the better, such as missing rate, outlier rate) Reverse standardization: ,in Avoid dividing by zero for extremely small positive numbers.

[0051] Logarithmic transformation can be enabled for heavy-tailed distributions: .

[0052] Missing data processing definition: In this application, missing data is strictly classified into two types: "Data gaps": These refer to data omissions during the data acquisition process where data values ​​that should have been present and recorded were not successfully acquired due to system failures, transmission interruptions, or missed acquisition points. Such omissions reflect quality issues in the data acquisition process.

[0053] "Undefined": refers to null values ​​that arise naturally due to the rationality of business logic, such as operational data not generated during equipment downtime, or null values ​​during non-working hours. These types of null values ​​are reasonable gaps at the business level.

[0054] Solution: For "missing" data, it is included in the inverse indicator of the integrity dimension for calculation. Specifically, the integrity performance is quantified by calculating the missing data rate; the higher the missing rate, the lower the integrity score. The calculation formula is: Missing rate = (Number of missing data points / Total number of data points to be collected) × 100%.

[0055] For "undefined" data, it is excluded from the denominator when calculating various indicators to ensure that the evaluation results are not negatively affected by null values ​​related to business rationality.

[0056] Abnormal data processing definition: Outlier data refers to data points that significantly deviate from the normal range due to measurement errors, transmission interference, equipment malfunctions, or other reasons. These outliers can severely impact the accuracy of data quality assessments.

[0057] Detection method: Anomaly detection employs a variety of statistical and machine learning methods: IQR (interquartile range) method: Based on the statistical characteristics of data distribution, data that exceeds the range of [Q1-1.5IQR, Q3+1.5IQR] are marked as outliers.

[0058] Three-Sigma criterion: For data that is approximately normally distributed, data points that exceed the range of μ±3σ are considered outliers.

[0059] Isolation Forest: It quickly identifies outliers that are significantly different from other data by constructing isolation trees. It is suitable for high-dimensional data and non-linear distributions.

[0060] Solution: After detecting outlier data, robust statistical methods are used for processing: Using the median instead of the mean for data aggregation reduces the impact of outliers on the overall statistics.

[0061] The mean is truncated, and a certain percentage (e.g., 5%) of extreme values ​​are removed before calculating the average.

[0062] Meanwhile, the outlier rate is an important input for the accuracy and consistency sub-indicators: Outlier rate = (number of outlier data points / total number of data points) × 100%.

[0063] This is a third type of mapping function that awards the highest score when the value is within the target range and decreases the score when it deviates from the range. This third type of mapping function is suitable for range-type indicators such as process parameters and equipment operating status. It uses a triangular or bell-shaped mapping, awarding full marks within the target range and decreasing the score linearly or Gaussianly based on distance outside the range.

[0064] Range / deviation type indicators (such as ± tolerance) Triangle / bell-shaped mapping: target interval Inside Outside the interval, the decay is linear or Gaussian based on distance:

[0065] This is the tolerance coefficient, used to control the smoothness of the descent curve.

[0066] The tolerance coefficient, used in the triangular / bell map, controls the sensitivity of the score to decay as it deviates from the target range. Its value directly determines the steepness of the score curve. The larger the coefficient, the slower the score decays, indicating a higher tolerance for deviation; the smaller the coefficient, the more rapid the decay, representing a lower tolerance for deviation.

[0067] Step 120-3: Calculate the score value of the sub-index according to the target mapping function.

[0068] This application's embodiments uniformly transform raw data with different dimensions into standard scores within the range of 0-100, establishing a unified quantitative benchmark.

[0069] Indicator Score: The This refers to the score of the i-th sub-index in the j-th dimension.

[0070] For example, for the inverse indicator of missing data rate, when the actual missing rate is 1.5%, the upper limit of the target is 2.5%, and the lower limit of the target is 0.5%, the score calculated by the second type of mapping function is 50 points.

[0071] Through this series of standardized quantitative processes, raw data from different data sources with different physical meanings were successfully transformed into comparable standard scores, providing a unified foundation for subsequent weight aggregation and comprehensive evaluation. This processing method effectively overcomes the problem in existing technologies where differences in indicator dimensions and distributions prevent direct quantification within the same scoring system, significantly improving the accuracy and reliability of evaluation results.

[0072] Step 130: According to the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

[0073] The weighting configuration refers to the relative importance coefficients assigned to each quality dimension. This definition clarifies the core role of weights in quality evaluation, namely, reflecting the differences in the criticality of different quality dimensions in a specific industrial scenario. The weighted aggregation refers to the process of linearly combining the quantitative values ​​of each quality dimension according to their weights. This process is a key step in integrating multi-dimensional quality evaluations into a single comparable score, providing a quantitative basis for the overall assessment of data quality.

[0074] Dimension weights: .in This represents the dimensional weight of the j-th dimension.

[0075] Dimensions and Comprehensive Aggregation Dimensional Score:

[0076] Overall Score:

[0077] This application introduces the Analytic Hierarchy Process (AHP) as a scientific method for determining weights. Through mathematical processes such as constructing judgment matrices and calculating eigenvectors, expert experience is transformed into objective weight allocation, effectively overcoming the problems of strong subjectivity and lack of mathematical basis in traditional methods for determining weights.

[0078] For example, in a typical configuration, the weights of the five dimensions—completeness, accuracy, consistency, timeliness, and availability—can be set to 0.20, 0.25, 0.20, 0.20, and 0.15, respectively. This configuration reflects the relative importance of the accuracy dimension in industrial data quality evaluation while maintaining a reasonable weight distribution for the other dimensions.

[0079] It should be noted that the typical configuration of the dimension weights described in this application: 0.20 / 0.25 / 0.20 / 0.20 / 0.15 is only one example, and the dimension weights can be adaptively learned and adjusted according to industry.

[0080] The system dynamically adjusts and calibrates dimensional weights based on different industries, scenarios, and data characteristics. It can also combine historical evaluation results, industry standards, and sample data from typical enterprises to differentiate and correct initial weights, better reflecting the actual importance of each dimension in different industrial sectors.

[0081] For example, in the process manufacturing industry, the timeliness and continuity of data are more critical, and the system will increase the weight of relevant dimensions. In contrast, in the discrete manufacturing industry, the weight of consistency and integrity dimensions can be appropriately strengthened to adapt to the requirements of data integration between equipment and process traceability. Through the above mechanism, this invention can achieve configurability, calibrability, and transferability of dimension weights, ensuring the scientific validity and applicability of the evaluation system in multiple industries and scenarios.

[0082] In one embodiment, step 130 determines that the weight configuration includes a consistency check; The weight configuration is effective when the consistency ratio (CR) is less than a set threshold.

[0083] This consistency detection mechanism is a key innovation of this application. It verifies the logical consistency of expert judgments using mathematical methods, ensuring the scientific validity and reliability of the weight allocation. The set threshold is typically 0.1. This empirical value is based on rigorous mathematical theory and effectively balances the rigor of the judgment with practical operability.

[0084] For example, construct a two-level judgment matrix based on three expert groups: "production / IT operations / quality management" (expert weights are recommended to be 4:3:3), and calculate the dimensional weights of the eigenvectors. With sub-indicator weights Consistency ratio It will take effect; otherwise, it will be returned for reassessment.

[0085] Further refinement: A hybrid weighting approach of "entropy weighting-analysis hierarchy" can be introduced, or empirical weights can be obtained by "regression calibration" using historical itemsets, with AHP used for fine-tuning to improve objectivity and transferability.

[0086] When the consistency ratio (CR) is less than a set threshold, the system determines that the weight configuration has passed the consistency check and can be used for subsequent weighted aggregation calculations. When the CR is greater than or equal to the set threshold, the judgment matrix needs to be readjusted until the consistency requirements are met. This mechanism effectively ensures the quality of the weight configuration and overcomes the problems of strong subjectivity and lack of objective and consistent mathematical verification methods in the weight determination process of existing technologies, making the evaluation results more scientific and credible.

[0087] For example, in determining the weight allocation, a hybrid weighting mechanism of "entropy weighting-analytic hierarchy process" can be introduced. Specifically, firstly, the prior distribution of empirical weights is obtained through regression calibration based on historical project datasets, and then fine-tuned using the analytic hierarchy process (AHP). This weighting method, which combines subjective and objective approaches, utilizes both the statistical regularities of historical data and incorporates expert understanding of the current scenario, significantly improving the objectivity of weight allocation and the transferability of the method across different industrial scenarios.

[0088] The comprehensive score generated through weighted aggregation not only reflects the overall level of data quality but also embodies the relative importance of each quality dimension, providing a unified metric for data quality comparison across systems and enterprises. This approach effectively addresses the lack of interpretability in existing technologies' comprehensive scores and graded ratings, and the difficulty for enterprises to identify specific shortcomings from the results, providing clear guidance for data quality governance.

[0089] In one embodiment, before the weighted aggregation of quantized values, the following steps are included: The quantified value is corrected based on the confidence factor of the data source and / or the data sample size.

[0090] When the data sample size is small or the collection confidence level is insufficient, the system automatically reduces the weight of the indicator to reduce random interference. The anomaly detection module can identify extreme data based on the Isolation Forest algorithm or the IQR method and perform robust corrections to the scoring results.

[0091] The confidence factor is a correction coefficient used to reflect the reliability of the data source. Its value is determined based on factors such as the calibration status of the data acquisition equipment, the stability of the transmission link, and the reliability level of the data source system. For example, data from metrologically certified high-precision sensors can be assigned a higher confidence factor, while data from uncalibrated portable devices are assigned a lower confidence factor. This correction mechanism can effectively distinguish data sources with different levels of reliability, avoiding excessive influence of low-reliability data on the evaluation results.

[0092] Confidence correction and sample size penalty Give each Overlaying data source confidence level , and sample size score :

[0093] The corrected score is obtained by adjusting the confidence level and sample size based on the index score S.

[0094] The data sample size correction is an adjustment of the quantified values ​​based on statistical principles, with the core purpose of addressing the instability of statistical results in the case of small samples. The correction process employs a monotonically increasing function of the sample size; when the actual sample size is less than the expected minimum sample size, an appropriate penalty adjustment is made to the quantified values. This design ensures that the evaluation results are not distorted by accidental statistical biases caused by small samples.

[0095] This embodiment establishes a dual safeguard mechanism by introducing a confidence factor and sample size correction, significantly improving the robustness and reliability of the evaluation system. In practice, the corrected quantified value is obtained by multiplying the original quantified value, the confidence factor, and the sample size correction coefficient. For example, when the original quantified value of an indicator is 80 points, the confidence factor is 0.9, and the sample size correction coefficient is 0.85, the final corrected value is 61 points.

[0096] This correction method effectively addresses the shortcomings of existing technologies that neglect the reliability of data sources and differences in sample size. It suppresses noise interference from low-quality data sources through confidence correction and avoids random biases in small sample statistics through sample size penalty. Thus, while ensuring interpretability, it significantly improves the stability and credibility of the comprehensive score, providing a more reliable quality assessment basis for data-driven decision-making in industrial enterprises.

[0097] Figure 3 A flowchart for determining the quality level of an industrial data quality quantitative evaluation method provided in this application includes steps 310-340.

[0098] This embodiment is in Figure 1 Based on the basic process shown, a quality level assessment and visualization analysis step has been added, establishing a complete mapping from quantitative scoring to level classification, which effectively solves the problem of poor interpretability of evaluation results in existing technologies.

[0099] Step 310: Obtain multidimensional data from industrial data sources.

[0100] This step corresponds to step 110, but in this embodiment, special emphasis is placed on obtaining a complete dataset for grading from various industrial data sources to ensure the comprehensiveness and representativeness of the evaluation basis.

[0101] Step 320: Quantize the multidimensional data according to multiple preset quality dimensions to determine the quantization value corresponding to each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values.

[0102] This step corresponds to step 120, but in this embodiment, the result of the quantification process is not only used for comprehensive score calculation, but more importantly, it provides a reliable quantitative basis for the grade determination in step 340, ensuring the scientific nature of the grade division.

[0103] Step 330: According to the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

[0104] This step corresponds to step 130, but in this embodiment, the calculation of the comprehensive score particularly emphasizes the scientific nature of the weight allocation, and uses weight values ​​that have passed the consistency test to ensure that the final score can truly reflect the overall level of data quality.

[0105] In one embodiment, the step of: Step 340: Compare the score with the preset grade threshold to determine the quality grade.

[0106] The preset grade thresholds are quality grade classification standards pre-defined based on industry standards and practical experience. These standards divide continuous comprehensive scores into several grade intervals with clear quality meanings. For example, four grades are set: Excellent, Good, Needs Improvement, and Unsatisfactory, each corresponding to a different score interval. This grading method transforms abstract data quality scores into intuitive quality grades, greatly enhancing the understandability and operability of the evaluation results.

[0107] The process of determining the quality level is achieved by comparing the overall score with a preset level threshold. Specifically, the system automatically matches the calculated overall score with the score range of each level to determine its corresponding quality level. For example, when the overall score is 85 points, according to the preset level threshold, the system will determine that it belongs to the "good" level.

[0108] This embodiment introduces a quality rating mechanism, establishing a mapping relationship from quantitative scores to quality levels, effectively solving the problem of insufficient interpretability in comprehensive scores in existing technologies. Simultaneously, the system outputs comparative analysis of quantitative values ​​for each quality dimension, generating visual charts such as radar charts. This multi-dimensional result presentation method enables enterprises not only to understand the overall level of data quality but also to accurately identify specific data quality shortcomings.

[0109] This approach, combining rating assessment with visualization analysis, provides enterprise management with a clear and intuitive view of data quality. This allows evaluation conclusions to be directly translated into targeted governance actions, effectively supporting continuous improvement of data quality. By transforming abstract scores into concrete levels and visual charts, this invention significantly enhances the practical value of evaluation results, providing reliable technical support for industrial enterprises to advance data asset management.

[0110] For example, a typical threshold: excellent: ; good:

[0111] Needs improvement:

[0112] Unqualified:

[0113] Specifically, after calculating the overall score, the system automatically generates a quality level report according to preset level thresholds, presenting the scores for different dimensions in the form of radar charts, bar charts, or cross-sectional analysis charts. This report can be simultaneously output to the enterprise data quality management platform or industrial internet portal, enabling management-level visual monitoring and early warning linkage. It also outputs radar charts and a list of shortcomings for each dimension, supporting "contribution / influence" analysis (Shapley or sensitivity approximation) for single indicators. Furthermore, it can generate a contribution and influence matrix for each indicator to the overall score through Shapley value decomposition, sensitivity analysis, or feature importance calculation methods. This analysis provides the data governance department with a prioritization basis, guiding the enterprise to rationally allocate governance resources and improvement investments. Further, the system supports automated report generation, outputting data quality diagnostic reports on a daily, weekly, or monthly basis, and pushing them to the quality management system or dashboard via API interface. This promotes the evolution of data quality from static monitoring to dynamic optimization.

[0114] Figure 4 A flowchart of a method for quantitative evaluation of industrial data quality and feedback governance suggestions is provided for embodiments of this application, including steps 410-450.

[0115] This embodiment is in Figure 1 Based on the basic process shown, a step of generating quality deficiency diagnosis and governance suggestions has been added, forming a complete evaluation-diagnosis-governance closed loop, which effectively solves the problem that evaluation results are difficult to translate into governance actions in existing technologies.

[0116] Step 410: Obtain multidimensional data from industrial data sources.

[0117] This step corresponds to step 110, but in this embodiment, special emphasis is placed on obtaining raw data for quality diagnosis from multiple sources such as MES, SCADA, energy and environmental systems, quality inspection systems, and logs, so as to provide a comprehensive data foundation for subsequent bottleneck analysis.

[0118] Step 420: Quantize the multidimensional data according to multiple preset quality dimensions to determine the quantization value corresponding to each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values.

[0119] This step corresponds to step 120, but in this embodiment, the result of the quantification process is not only used for comprehensive score calculation, but more importantly, it provides detailed scoring basis for each dimension for the identification of quality shortcomings in step 440.

[0120] Step 430: According to the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

[0121] This step corresponds to step 130. The generated comprehensive score is used to determine the overall quality level and provides a reference benchmark for the relative performance of each dimension.

[0122] In one embodiment, the step of: Step 440: Compare the quantified values ​​of multiple quality dimensions with the preset quality thresholds.

[0123] The preset quality thresholds are qualification standards for each dimension based on industry best practices and business requirements, typically expressed as the minimum acceptable score for each dimension. For example, the quality threshold for the completeness dimension might be set at 70 points, and the quality threshold for the accuracy dimension might be set at 80 points. This comparison process can systematically identify which quality dimensions have significant weaknesses that require priority addressing.

[0124] Step 450: In response to a quantization value being less than the corresponding quality threshold, determine the data governance recommendations corresponding to the quantization value based on a preset rule base.

[0125] The preset rule base is a knowledge base that stores the mapping relationship between quality issues and governance actions, and contains specific governance solutions for different types of quality issues.

[0126] For example, when the quantified value of the integrity dimension is lower than the quality threshold, the system will suggest checking the status of the data collection interface based on the rule base and supplementing the data for the missing periods. When the quantified value of the consistency dimension is lower than the quality threshold, the system will suggest verifying and unifying the data units and formats from different systems and revising the data access specifications. This targeted suggestion generation mechanism transforms abstract quality evaluation results into concrete and actionable governance actions, greatly improving the efficiency and effectiveness of data quality governance.

[0127] The rule base can be established based on historical governance experience, industry standards, and typical problem templates. The system can automatically generate governance strategies, such as "optimization of supplementary collection mechanism", "adjustment of data caliber", "correction of sampling frequency" and "cleaning of interface cache".

[0128] For example, the preset rule base can be an industry indicator template base. The industry indicator template base is a set of standardized indicators pre-set based on the characteristics of different industrial sectors. While maintaining the five core dimensions of completeness, accuracy, consistency, timeliness and availability, it configures corresponding sub-indicator sets and target value bases for the business characteristics and data features of different industries.

[0129] To enable cross-industry application of this method, the system provides a configurable "industry indicator template library" covering typical industrial scenarios such as steel, mining, chemical, and power. The core of this mechanism lies in allowing users to dynamically replace the sub-indicator sets and corresponding target value libraries under each dimension based on industry characteristics, without altering the framework of the five core dimensions of "completeness, accuracy, consistency, timeliness, and usability."

[0130] Target values ​​can be set from various sources, including: 1) national regulations and industry standards; 2) Service Level Agreement (SLA) requirements; and 3) quantile (P90, P95) statistical results of historical data. These target values ​​can be configured hierarchically according to "equipment type", "process section", "work group", "region", etc., thereby achieving refined benchmarking management between different factories and production lines within the same group and accurately identifying quality shortcomings.

[0131] Implementation method: Establish indicator templates covering typical industrial sectors such as steel, mining, chemicals, and power.

[0132] Each industry template includes industry-specific Key Quality Indicators (KQI) and Data Quality Indicators (DQI).

[0133] It supports hierarchical configuration based on multiple dimensions such as equipment type, process section, work group, and region.

[0134] Figure 5 This is a structural diagram of an industrial data quality quantitative evaluation device according to an embodiment of this application, used to implement the industrial data quality quantitative evaluation method described in any embodiment of the first aspect, including: The acquisition module 501 is used to acquire multidimensional data from industrial data sources.

[0135] The determination module 502 is used to quantize the multidimensional data according to multiple preset quality dimensions and determine the quantization value corresponding to each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values.

[0136] The calculation module 503 is used to weight and aggregate the quantified values ​​of multiple quality dimensions according to a preset weight configuration to generate a comprehensive score.

[0137] Furthermore, the acquisition module includes a first acquisition unit for acquiring multidimensional data from industrial data sources.

[0138] The determining module includes a first determining unit, used to quantize the multidimensional data according to multiple preset quality dimensions and determine the quantized value corresponding to each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless scoring values.

[0139] The calculation module includes a first calculation unit, which is used to weight and aggregate the quantified values ​​of multiple quality dimensions according to a preset weight configuration to generate a comprehensive score.

[0140] In one embodiment, the determining module further includes a second determining unit for identifying the data orientation of sub-indicators corresponding to the quality dimension in the multidimensional data; the data orientation includes positive indicators, negative indicators, or interval indicators.

[0141] The determining module further includes a third determining unit, used to determine a preset target mapping function based on the identified data orientation.

[0142] The calculation module further includes a second calculation unit for calculating the score value of the sub-index according to the target mapping function.

[0143] The above embodiments are used to implement the technical features described in the specification regarding identifying data orientation, determining the mapping function, and calculating the score value.

[0144] In one embodiment, the determining module further includes a fourth determining unit, used to determine the weight configuration using the analytic hierarchy process (AHP) and to perform a consistency check on the weight configuration; in response to the consistency ratio (CR) value being less than a set threshold, the weight configuration is confirmed to be valid.

[0145] The above embodiments are used to implement the technical features described in the specification regarding the use of the AHP method for weight determination and consistency verification.

[0146] In one embodiment, the calculation module further includes a correction unit for correcting the quantized values ​​according to the confidence factor of the data source and / or the data sample size before the quantized values ​​are weighted and aggregated.

[0147] The above embodiments are used to implement the technical features described in the specification regarding the correction of quantized values ​​based on confidence factors and sample size.

[0148] In one embodiment, the device further includes a grade determination module for comparing the score value with a preset grade threshold to determine the quality grade.

[0149] The above embodiments are used to implement the technical feature in the specification regarding determining the quality level by comparing the score value with the level threshold.

[0150] In one embodiment, the determining module further includes a fifth determining unit for determining the target mapping function, which includes: a first type of mapping function where the larger the value, the higher the score; a second type of mapping function where the smaller the value, the higher the score; and a third type of mapping function where the score is highest when the value is within the target range and decreases when it deviates from the range.

[0151] The above embodiments are used to implement the technical features regarding the definition and use of the three types of mapping functions in the specification.

[0152] In one embodiment, the device further includes a diagnostic suggestion module for comparing quantified values ​​of multiple quality dimensions with preset quality thresholds; and in response to a quantified value being less than the corresponding quality threshold, determining a data governance suggestion corresponding to the quantified value based on a preset rule base.

[0153] The above embodiments are used to implement the technical features for identifying and generating suggestions for addressing quality shortcomings as described in the specification.

[0154] The industrial data quality quantitative evaluation device further realizes systematic functions through the following modules: Data Access and Governance Module: Responsible for connecting to heterogeneous industrial data from multiple sources such as MES, SCADA, energy and environmental protection, quality inspection, and logs, completing the unification of data standards and alignment with master data, and providing a clean and consistent data source for evaluation.

[0155] Indicator Engine: Based on predefined indicator metadata (including direction, type, unit, threshold, etc.), it automatically selects the appropriate family of quantitative mapping functions and completes the calculation of indicator scores.

[0156] Weighting Engine: Supports multiple weighting algorithms such as Analytic Hierarchy Process (AHP), Entropy Weighting, and Regression Calibration, and has a built-in Consistency Ratio (CR) test function to ensure that the weights are scientific and effective.

[0157] Scoring and Grading Module: Responsible for calculating the overall score and quality level, and automatically generating visual analysis results such as radar charts, cross-sectional diagrams, and lists of shortcomings.

[0158] The explanation and suggestion module automatically matches and generates specific data governance suggestions from the preset rule base for the "shortcoming indicators" identified in the evaluation results. These suggestions include: supplementing data collection points, modifying interfaces, unifying standards, synchronizing clocks, and adjusting caches and queues, forming a closed loop of "evaluation-diagnosis-governance".

[0159] Configurable portal and API: Provides a user configuration interface and application programming interface, supporting the orchestration, version management, audit traceability, and integration with external business systems for evaluation tasks.

[0160] Device / Media: All the above functions can be implemented by the processor executing instructions stored in the storage medium, supporting hybrid deployment on local servers, cloud platforms, or edge computing nodes to meet the diverse IT architecture needs of industrial enterprises. Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] Therefore, this application also proposes a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the methods described in any embodiment of this application.

[0162] Furthermore, this application also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any embodiment of this application.

[0163] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0164] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory. Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0167] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 600 shown is merely an example and should not impose any limitations on the function and scope of use of the embodiments of this application. It includes: one or more processors 620; and a storage device 610 for storing one or more programs, which, when run by the one or more processors 620, cause the one or more processors 620 to implement the industrial data quality quantitative evaluation and determination method provided in the embodiments of this application, the method including: Acquire multidimensional data from industrial data sources; The multidimensional data is quantized according to multiple preset quality dimensions to determine the quantized value for each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values. Based on the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

[0168] The electronic device 600 also includes an input device 630 and an output device 640; the processor 620, storage device 610, input device 630 and output device 640 in the electronic device can be connected by a bus or other means, as shown in the figure, which is connected by a bus 650.

[0169] Storage device 610, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and module units, such as the program instructions corresponding to the industrial data quality quantitative evaluation method in this embodiment. Storage device 610 may mainly include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on terminal usage. Furthermore, storage device 610 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, storage device 610 may further include memory remotely located relative to processor 620, and these remote memories can be connected via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0170] Input device 630 can be used to receive input digital, character, or voice information, and to generate key signal inputs related to user settings and function control of the electronic device. Output device 640 may include electronic devices such as a display screen and a speaker.

[0171] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0172] Those skilled in the art will understand that, unless otherwise stated, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be understood that when a device or component is “connected” to another device or component, it may be directly connected to the other device or component, or there may be an intermediary device or component. Furthermore, the term “connection” as used herein may include partially wireless connections as well as partially wired connections.

[0173] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0174] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for quantitatively evaluating the quality of industrial data, characterized in that, Including the following steps: Acquire multidimensional data from industrial data sources; The multidimensional data is quantized according to multiple preset quality dimensions to determine the quantized value for each quality dimension; the quantization process includes mapping the original data with different dimensions to dimensionless score values. Based on the preset weight configuration, the quantified values ​​of multiple quality dimensions are weighted and aggregated to generate a comprehensive score.

2. The method for quantitative evaluation of industrial data quality according to claim 1, characterized in that, The quantization process specifically includes: Identify the data orientation of sub-indicators corresponding to the quality dimension in the multidimensional data; the data orientation includes positive indicators, negative indicators, or interval indicators; Based on the identified data orientation, a preset target mapping function is determined; The score value of the sub-index is calculated based on the target mapping function.

3. The method for quantitative evaluation of industrial data quality according to claim 1, characterized in that, Determining the weight configuration includes consistency checks; The weight configuration is effective when the consistency ratio (CR) is less than a set threshold.

4. The method for quantitative evaluation of industrial data quality according to claim 1, characterized in that, Before the weighted aggregation of quantified values, the following steps are included: The quantified value is corrected based on the confidence factor of the data source and / or the data sample size.

5. The method for quantitative evaluation of industrial data quality according to claim 1, characterized in that, It also includes the following steps: The quality level is determined by comparing the score with a preset level threshold.

6. The method for quantitative evaluation of industrial data quality according to claim 2, characterized in that, The target mapping function includes: The higher the value, the higher the score of the first-class mapping function; The second type of mapping function awards higher scores for smaller values. The third type of mapping function is one in which the score is highest when the value is within the target range and decreases when it deviates from the range.

7. The method for quantitative evaluation of industrial data quality according to claim 1, characterized in that, It also includes the following steps: Compare the quantified values ​​of multiple quality dimensions with preset quality thresholds; In response to a quantification value being less than the corresponding quality threshold, a data governance recommendation corresponding to the quantification value is determined based on a preset rule base.

8. An industrial data quality quantitative evaluation device, used to implement the industrial data quality quantitative evaluation method according to any one of claims 1-7, characterized in that, include: The acquisition module is used to acquire multidimensional data from industrial data sources; The determination module is used to quantize the multidimensional data according to multiple preset quality dimensions and determine the quantization value corresponding to each quality dimension; the quantization process includes uniformly mapping the original data with different dimensions to dimensionless score values. The calculation module is used to weight and aggregate the quantified values ​​of multiple quality dimensions according to a preset weight configuration to generate a comprehensive score.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

10. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.

Images