Quality management system evaluation and authentication system based on multi-source heterogeneous data processing

Abstract

The invention discloses a quality management system evaluation and authentication system based on multi-source heterogeneous data processing, and belongs to the technical field of quality management. A quality management system evaluation and authentication system based on multi-source heterogeneous data processing comprises a data acquisition module, a data preprocessing module, a data fusion module, a data analysis module, a quality management system grading model, an evaluation and authentication module and a user management and interaction module. According to the invention, the problems of single data source, need of manual acquisition of evaluation data and poor evaluation efficiency in evaluation authentication of the existing quality management system are solved. Through automatic acquisition and processing of the multi-source heterogeneous data, the comprehensiveness and accuracy of the data are remarkably improved, the labor cost is greatly reduced, the evaluation efficiency and quality are improved, and a more reliable quality management evaluation result is provided for enterprises.

Description

Technical Field

[0001] This invention relates to the field of quality management technology, specifically a quality management system assessment and certification system based on multi-source heterogeneous data processing. Background Technology

[0002] Quality management system assessment and certification is a crucial means of ensuring product quality and operational efficiency for enterprises. However, traditional quality management system assessment and certification processes suffer from several significant problems that limit the comprehensiveness and efficiency of the assessment. These problems mainly include the following:

[0003] First, the data sources are limited. The data sources for traditional quality management system assessment and certification mainly rely on internal enterprise data, such as production records, quality inspection reports, and employee training records. Although these data can provide a certain level of quality management information, they are often limited to internal enterprise data and lack external data support. Some important external information is not fully incorporated into the assessment system, and the system cannot fully reflect the quality management status of the enterprise.

[0004] Secondly, manual data collection is crucial in traditional assessment processes. A large amount of data needs to be collected manually, which is time-consuming, inefficient, prone to errors, and makes it difficult to guarantee the timeliness and accuracy of the data. Furthermore, it requires a significant amount of human resources for data collection and processing, increasing the company's operating costs. Different personnel may use different standards and methods when collecting data, resulting in poor data consistency and affecting the reliability of the assessment results. Summary of the Invention

[0005] The purpose of this invention is to provide a quality management system assessment and certification system based on multi-source heterogeneous data processing. Through automatic collection and processing of multi-source heterogeneous data, the system significantly improves the comprehensiveness and accuracy of the data, greatly reduces labor costs, and improves assessment efficiency and quality, providing enterprises with more reliable quality management assessment results and solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A quality management system assessment and certification system based on multi-source heterogeneous data processing includes:

[0008] The data acquisition module, based on the industry type of the company being evaluated, clarifies the types of data and data sources to be collected, and automatically collects information from multiple data sources through a data interface;

[0009] The data preprocessing module performs data cleaning, format conversion, missing data imputation, and outlier handling to ensure data consistency.

[0010] The data fusion module integrates data from different sources to form a unified data view;

[0011] The data analysis module uses data analysis techniques to perform in-depth analysis on the preprocessed data, and the data analysis results are used to build a hierarchical model for the quality management system.

[0012] A tiered quality management system model is constructed based on the core evaluation module of the effectiveness of the quality management system and in conjunction with product quality levels.

[0013] The assessment and certification module defines clear assessment indicators to comprehensively evaluate a company's quality management system and provide certification results.

[0014] The user management and interaction module is used to assign user roles, and users can perform data management and evaluation / authentication management within the scope of their assigned permissions.

[0015] Preferably, the data acquisition module includes:

[0016] Automatic data acquisition interface unit, used for:

[0017] Connect to various internal management systems of the enterprise through API interfaces or databases, including at least ERP system, CRM system, SCM system, QMS system, R&D system, production system and human resources system;

[0018] Connect to external data sources via file transfer protocols to obtain data from supplier systems, acquire audit reports from third-party certification bodies via file download, and obtain industry standards and legal requirements from the network via API interfaces;

[0019] Connect to social media platforms via APIs to obtain tweets, user comments, browsing data, and interaction data about the company's products and services;

[0020] The data acquisition engine unit is used for:

[0021] Schedule data acquisition tasks according to the configured timetable or triggering conditions;

[0022] Write an adapter for each data source to implement the specific data acquisition logic. The adapter is responsible for communicating with the data source and extracting the required data.

[0023] Manual data input unit, used for:

[0024] Data that cannot be automatically collected is supplemented and improved by manual input.

[0025] Preferably, the data preprocessing module further includes:

[0026] Data cleaning includes, but is not limited to, removing duplicate data, handling missing values, and correcting erroneous data;

[0027] Standardize the data, including but not limited to unifying date format, numerical format, text format, unit conversion, field mapping, and data normalization;

[0028] Perform quality checks, integrity checks, and accuracy checks on the data.

[0029] Preferably, the data fusion module is used for:

[0030] Merge data from different data sources that have the same records into a single table and use unique identifiers to associate them;

[0031] Data from different data sources but describing the same entity at different time periods is merged into a single table, and a timestamp field is introduced to mark the validity period of the data, supporting backtracking queries of historical data;

[0032] It links data from multiple tables to form a complete data view, using JOIN operations to achieve multi-table joins; it supports star and snowflake schema data warehouse architectures, and uses a rule base for custom rules and automated matching, while leveraging natural language processing technology to assist in field name parsing.

[0033] Connect related data tables to form a complete data system, and use ETL tools to extract, transform and load data;

[0034] The merged data is output to the designated data storage location for use in quality management system assessment and certification, and the output data format is ensured to meet the requirements of the quality management system assessment and certification system; the storage path and file format are configured to ensure compatibility with downstream analysis tools.

[0035] Preferably, the data analysis module is used for:

[0036] Using batch insertion and streaming loading mechanisms, data views and data systems are received from the data fusion module, and predefined patterns are used to verify whether the format and structure of the data meet the expected requirements.

[0037] Use NumPy to calculate descriptive statistics, including at least the mean, standard deviation, and distribution shape, and use the Z-score method to identify outliers, basic characteristics of the data, and potential problems.

[0038] The ARIMA model is applied to capture long-term trends and periodic fluctuations. Wavelet transform is used to decompose signal components and reveal hidden patterns. Time series invariance analysis is used to extract features from the data and obtain trend changes and interdependencies between multi-source data.

[0039] Choose the appropriate model based on the complexity and interpretability of the data. Use the random forest model for the classification and grading problem of quality management system, the random forest regression model for the product quality score prediction problem, and the isolation forest model for quality problem identification.

[0040] The model parameters were optimized using a grid search method, and the tuning process was accelerated using Bayesian optimization. The model was evaluated and certified using accuracy, recall, and F1 score metrics.

[0041] For changes and reorganizations of initial data, apply either correlation analysis or principal component analysis to select the most influential features and create composite features to form new feature variables;

[0042] The extracted features are used to train machine learning models to uncover potential patterns and relationships between data, and the model performance is evaluated and optimized through cross-validation.

[0043] Preferably, the quality management system hierarchical model is used for:

[0044] Based on the relationship between the quality management system and product quality work, the core evaluation module that reflects the effectiveness of the quality management system is determined. Based on the data and information provided by the data analysis module, specific evaluation standards and a grading system are defined, and a grading model of the organization's quality management system is constructed.

[0045] The quality management system classification model includes five levels: A, B, C, D, and E.

[0046] Preferably, the characteristics of the quality management system level are as follows:

[0047] The characteristics of Grade A quality management are: quality management activities are usually ad hoc and lack standardization and documentation;

[0048] The characteristics of Grade B quality management are: the establishment of basic quality management processes, with initial documentation;

[0049] The characteristics of Level C quality management are: the quality management process has been standardized and documented, and employees have received training and follow the quality management process;

[0050] The characteristics of Level D quality management are: the quality management process is not only standardized, but also measured and controlled to ensure consistency and efficiency;

[0051] The quality management characteristics of Level E are: the quality management process is not only measured and controlled, but also continuously optimized through continuous improvement and innovation.

[0052] Preferably, the workflow of the evaluation and certification module includes the following steps:

[0053] Specific evaluation indicators are defined according to the quality management system grading model. These evaluation indicators include, but are not limited to, the degree of documentation of the quality management process, the status of quality training, the frequency of quality audits, the frequency of quality data collection and analysis, and the mechanisms and records for continuous quality improvement.

[0054] This involves transforming qualitative evaluation indicators into quantitative evaluation standards;

[0055] Develop scoring criteria for each evaluation indicator and quantify the score for each indicator;

[0056] The importance of each indicator in the overall evaluation is determined by the analytic hierarchy process, weights are assigned to each indicator, and the total score is calculated.

[0057] Preferably, the workflow of the evaluation and certification module further includes the following steps:

[0058] The company's quality management level is classified into the corresponding quality management system level based on the total score;

[0059] The preliminary assessment results were reviewed to confirm the completeness and accuracy of the data. Any problems found were recorded, and suggestions for improvement were put forward.

[0060] Items that do not meet the requirements will be rejected, marked, and comprehensively evaluated to determine whether they will affect the final evaluation results.

[0061] Summarize the results of various assessments, conduct a comprehensive analysis, and finally determine the level of the quality management system based on the comprehensive evaluation results;

[0062] Generate a final evaluation report and provide a certification certificate;

[0063] Explain the assessment results to the company, propose specific quality management improvement suggestions based on the assessment results, and develop an improvement plan.

[0064] Preferably, the user management and interaction module includes:

[0065] Enterprise management unit, used for:

[0066] Please upload additional documents for the quality management assessment.

[0067] View the assessment progress in real time and understand the current assessment stage;

[0068] Receive the final evaluation and certification results;

[0069] View the detailed assessment report and improvement recommendations;

[0070] The assessment and certification management unit is used for:

[0071] Configure the data acquisition task, including at least the acquisition frequency, data source, and acquisition rules;

[0072] Monitor data acquisition status in real time and view data acquisition logs;

[0073] Receive notifications of anomalies during the data acquisition, data preprocessing, data fusion, data analysis, and evaluation and certification processes;

[0074] Review and approve the materials submitted by the company;

[0075] Update the assessment progress and record the completion status of each assessment and certification stage.

[0076] Preferably, the user management and interaction module further includes:

[0077] The user management unit is used for:

[0078] Supports registration and login for both businesses and assessment / certification bodies;

[0079] Different permissions are assigned based on user roles.

[0080] Compared with the prior art, the beneficial effects of the present invention are:

[0081] 1. This invention automatically collects data from multiple sources, including internal enterprise management systems, external data sources, and social media platforms. By integrating data from different sources, it provides more comprehensive and accurate evaluation results, helping enterprises gain deeper insights and thus better conduct quality management assessments and certifications.

[0082] 2. This invention utilizes multiple methods such as API interfaces, database connections, and file transfer protocols to automatically collect information from various data sources, reducing manual operations, improving the efficiency and accuracy of data collection, and lowering the cost of data collection. At the same time, it sets up a manual data input unit to supplement and improve data that cannot be automatically collected through manual input, ensuring that all necessary data can be included in the evaluation system and guaranteeing the comprehensiveness of the evaluation data.

[0083] 3. This invention combines the enterprise management unit and the assessment and certification management unit, allowing enterprises to upload materials and information for assessment, view assessment progress, and receive final certification results. It also allows assessment and certification parties to configure data collection tasks and rules and monitor data collection status, promoting collaboration between enterprises and assessment and certification parties. This facilitates enterprise participation in the assessment process, enabling them to understand assessment progress and results in a timely manner, thus improving the transparency and participation of the assessment. Furthermore, assessment and certification parties can effectively manage the assessment process, improving the efficiency and accuracy of the assessment. Attached Figure Description

[0084] Figure 1This is a system module structure diagram of the present invention;

[0085] Figure 2 This is a flowchart of the evaluation and certification process for this invention. Detailed Implementation

[0086] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0087] To address the issues of limited data sources, reliance on manual data collection, and low efficiency in existing quality management system assessment and certification processes, please refer to [link / reference]. Figure 1 and Figure 2 This embodiment provides the following technical solution:

[0088] A quality management system assessment and certification system based on multi-source heterogeneous data processing includes:

[0089] The data acquisition module, based on the industry type of the company being evaluated, clarifies the types of data and data sources to be collected, and automatically collects information from multiple data sources through a data interface;

[0090] The data preprocessing module performs data cleaning, format conversion, missing data imputation, and outlier handling to ensure data consistency.

[0091] The data fusion module integrates data from different sources to form a unified data view;

[0092] The data analysis module uses data analysis techniques to perform in-depth analysis on the preprocessed data, and the data analysis results are used to build a hierarchical model for the quality management system.

[0093] A tiered quality management system model is constructed based on the core evaluation module of the effectiveness of the quality management system and in conjunction with product quality levels.

[0094] The assessment and certification module defines clear assessment indicators to comprehensively evaluate a company's quality management system and provide certification results.

[0095] The user management and interaction module is used to assign user roles, and users can perform data management and evaluation / authentication management within the scope of their assigned permissions.

[0096] The data acquisition module includes:

[0097] Automatic data acquisition interface unit, used for:

[0098] Connect to various internal management systems of the enterprise through API interfaces or databases, including at least ERP system, CRM system, SCM system, QMS system, R&D system, production system and human resources system;

[0099] Connect to external data sources via file transfer protocols to obtain data from supplier systems, acquire audit reports from third-party certification bodies via file download, and obtain industry standards and legal requirements from the network via API interfaces;

[0100] Connect to social media platforms via APIs to obtain tweets, user comments, browsing data, and interaction data about the company's products and services.

[0101] By utilizing API interfaces, database connections, file transfer protocols, and other methods, information is automatically collected from multiple data sources, reducing manual operations, improving the efficiency and accuracy of data collection, and lowering the cost of data collection. At the same time, by obtaining external data from suppliers, third-party certification bodies, websites, and social media platforms through file transfer protocols, the data sources are enriched, enhancing the comprehensiveness of the assessment and certification of quality management.

[0102] The data acquisition engine unit is used to schedule data acquisition tasks according to the configured schedule or triggering conditions; it writes adapters for each data source to implement specific acquisition logic. The adapters are responsible for communicating with the data source, extracting the required data, ensuring that the data acquisition tasks are executed as planned, improving the reliability and consistency of data acquisition, enabling the system to flexibly adapt to the characteristics of different data sources, and improving the scalability of the system.

[0103] The manual data input unit supplements and improves data that cannot be automatically collected by manually inputting data, ensuring that all necessary data can be included in the evaluation system and avoiding incomplete or inaccurate evaluations due to missing data.

[0104] The data preprocessing module further includes: cleaning the data, including but not limited to removing duplicate data, handling missing values, and correcting erroneous data to improve data quality and ensure the accuracy of subsequent analysis; standardizing the data, including but not limited to unifying date formats, numerical formats, text formats, unit conversions, field mappings, and data normalization to ensure that data from different sources are consistent in format, facilitating subsequent data fusion and analysis; and performing quality checks, integrity checks, and accuracy checks on the data to ensure data integrity and accuracy, providing a reliable foundation for subsequent analysis.

[0105] The data fusion module is used for:

[0106] Data from different data sources but with the same records are merged into a single table and linked using unique identifiers. When handling conflicts, the latest or most authoritative data version is retained first.

[0107] Data from different data sources but describing the same entity at different time periods is merged into a single table, and a timestamp field is introduced to mark the validity period of the data, supporting backtracking queries of historical data;

[0108] This feature links data from multiple tables to form a complete data view, facilitating comprehensive data analysis and improving the comprehensiveness and accuracy of assessments. It uses JOIN operations to achieve multi-table joins, optimizing query statements for improved performance. It supports data warehouse architectures in both Star and Snowflake schemas, and utilizes a rule base for custom rules and automated matching, while leveraging natural language processing to assist in field name resolution and optimize query performance.

[0109] The related data tables are connected to form a complete data system, and ETL tools (such as Apache NiFi and Talend) are used for data extraction, transformation and loading; incremental updates are supported to reduce the resource consumption caused by full refresh;

[0110] The merged data is output to a specified data storage location for use in quality management system assessment and certification, ensuring that the output data format meets the requirements of the quality management system assessment and certification system; the storage path and file format are configured to ensure compatibility with downstream analysis tools; compression and encryption options are supported to ensure data transmission security.

[0111] The data analysis module is used for:

[0112] Using batch insertion and streaming loading mechanisms, data views and data systems are received from the data fusion module, and predefined patterns are used to verify whether the format and structure of the data meet the expected requirements.

[0113] Use NumPy to calculate descriptive statistics, including at least the mean, standard deviation, and distribution shape, and use the Z-score method to identify outliers, basic characteristics of the data, and potential problems.

[0114] The ARIMA model is applied to capture long-term trends and periodic fluctuations, wavelet transform is used to decompose signal components and reveal hidden patterns, and time series analysis is used to identify the interdependencies between multi-source data.

[0115] The initial data is modified and reorganized to form new feature variables to enhance model performance. Correlation analysis or principal component analysis is applied to calculate the correlation coefficient matrix between features, and the feature with the greatest impact is selected. Composite features are created by combining domain consensus, and feature construction rules are defined to ensure the rationality and interpretability of new features and avoid overfitting.

[0116] Considering the complexity and interpretability of the data, choose an appropriate model. For the classification and grading problem of quality management system, use the random forest model; for the product quality score prediction problem, use the random forest regression model; and for quality problem identification, use the isolation forest model.

[0117] The model parameters were optimized using a grid search method, the tuning process was accelerated using Bayesian optimization, and the model was evaluated and certified using accuracy, recall, and F1 score metrics.

[0118] The extracted features are used to train machine learning models to uncover potential patterns and relationships between data, and the model performance is evaluated and optimized through cross-validation.

[0119] The merged and analyzed data is output to the designated data storage location for use in quality management system assessment and certification, and the output data format is ensured to meet the requirements of the quality management system assessment and certification system.

[0120] The quality management system grading model further includes:

[0121] Based on the relationship between the quality management system and product quality work, a core evaluation module reflecting the effectiveness of the quality management system is identified. Based on the data and information provided by the data analysis module, specific evaluation standards and a grading system are defined. Combined with product quality levels, a grading model of the organization's quality management system is constructed. Through the grading model, the maturity of the company's quality management can be objectively evaluated, helping the company identify areas for improvement and gradually enhance its management level.

[0122] The quality management system grading model includes five levels: A, B, C, D, and E. The specific characteristics of each level are as follows: Level A: Quality management activities are typically ad hoc, lacking standardization and documentation; Level B: Basic quality management processes are being established and are initially documented; Level C: Quality management processes are standardized and documented, and employees are trained and follow them; Level D: Quality management processes are not only standardized but also measured and controlled to ensure consistency and efficiency; Level E: Quality management processes are not only measured and controlled but also continuously optimized through continuous improvement and innovation. Understanding the characteristics of each level helps companies understand their current stage and develop targeted improvement measures.

[0123] The workflow for the assessment and certification module includes the following steps:

[0124] Specific evaluation indicators are defined according to the quality management system grading model. These evaluation indicators include, but are not limited to, the degree of documentation of the quality management process, the status of quality training, the frequency of quality audits, the frequency of quality data collection and analysis, and the mechanisms and records for continuous quality improvement.

[0125] This involves transforming qualitative evaluation indicators into quantitative evaluation standards;

[0126] Develop scoring criteria for each evaluation indicator and quantify the score for each indicator; determine the importance of each indicator in the overall evaluation through the analytic hierarchy process, assign weights to each indicator, and calculate the total score.

[0127] Based on the total score, the company's quality management level is classified into the corresponding quality management system level; the preliminary assessment results are reviewed to confirm the completeness and accuracy of the data, problems found are recorded, and improvement suggestions are proposed; non-compliant items are rejected, marked, and comprehensively evaluated to determine whether they affect the final assessment results;

[0128] Summarize all assessment results, conduct comprehensive analysis, and ultimately determine the level of the quality management system based on the comprehensive evaluation results; generate the final assessment report and provide certification; explain the assessment results to the company, propose specific quality management improvement suggestions based on the assessment results, and formulate improvement plans.

[0129] By comprehensively considering qualitative and quantitative indicators, the evaluation results are made more comprehensive and objective, improving the credibility of the evaluation; quantifying the evaluation results facilitates comparison and analysis, improving the transparency and fairness of the evaluation; review and improvement suggestions ensure the accuracy and reliability of the evaluation results, provide enterprises with specific improvement suggestions, and promote continuous improvement.

[0130] The user management and interaction module includes:

[0131] Enterprise management unit, used for:

[0132] Upload supplementary documents for quality management assessment; view assessment progress in real time and understand the current assessment stage; receive final assessment and certification results; view detailed assessment reports and improvement suggestions, facilitating enterprise participation in the assessment process, timely understanding of assessment progress and results, and improving the transparency and participation of the assessment.

[0133] The assessment and certification management unit is used for: configuring data collection tasks, including at least the collection frequency, data source, and collection rules; monitoring the data collection status in real time and viewing the data collection logs; receiving notifications of anomalies during the data collection, data preprocessing, data fusion, data analysis, and assessment and certification processes; viewing and reviewing the materials submitted by the company; updating the assessment progress and recording the completion status of each assessment and certification stage to ensure that the assessment and certification party can effectively manage the assessment process and improve the efficiency and accuracy of the assessment.

[0134] By combining the enterprise management unit and the assessment and certification unit, the system supports enterprises in uploading materials and information for assessment, viewing assessment progress, and receiving final certification results. It also supports assessment and certification units in configuring data collection tasks and rules and monitoring data collection status. This promotes collaboration between enterprises and assessment and certification units, facilitates enterprise participation in the assessment process, and allows them to understand assessment progress and results in a timely manner, thereby improving the transparency and participation of the assessment. Assessment and certification units can also effectively manage the assessment process, improving the efficiency and accuracy of the assessment.

[0135] The user management unit supports registration and login operations for enterprises and assessment / certification parties; different permissions are assigned according to user roles.

[0136] Working principle: Based on the industry type of the company being evaluated, the system identifies the types of data to be collected and the data sources. It automatically collects information from multiple sources, including the company's internal management system, external data sources, and social media platforms. The system performs data cleaning, format conversion, missing data imputation, and outlier handling to ensure data consistency and accuracy. It also standardizes the data to lay the foundation for subsequent data fusion and analysis. Finally, it integrates data from different sources to form a unified data view. Through data merging, association, and connection, it constructs a complete data system and outputs it to the designated data storage location for use in quality management system assessment and certification.

[0137] The system utilizes data analytics to deeply analyze preprocessed data, identifying trends and interdependencies among multiple data sources. Based on the core evaluation module for the effectiveness of the quality management system, and combined with product quality levels, it comprehensively assesses the quality management system using qualitative and quantitative methods, providing certification results. Specific evaluation indicators are defined, qualitative indicators are transformed into quantitative evaluation standards, scoring criteria are developed, a total score is calculated, and the company's quality management system level is ultimately determined. A detailed evaluation report is generated, providing specific improvement suggestions. Companies can upload materials and information for evaluation, view evaluation progress, and receive final certification results. The evaluation and certification body can configure data collection tasks and rules, and monitor data collection status.

[0138] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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 process, method, article, or apparatus.

[0139] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A quality management system assessment and certification system based on multi-source heterogeneous data processing, characterized in that: include: The data acquisition module, based on the industry type of the company being evaluated, clarifies the types of data and data sources to be collected, and automatically collects information from multiple data sources through a data interface; The data preprocessing module performs data cleaning, format conversion, missing data imputation, and outlier handling to ensure data consistency. The data fusion module integrates data from different sources to form a unified data view; The data analysis module uses data analysis techniques to perform in-depth analysis on the preprocessed data, and the data analysis results are used to build a hierarchical model for the quality management system. A tiered quality management system model is constructed based on the core evaluation module of the effectiveness of the quality management system and in conjunction with product quality levels. The assessment and certification module defines clear assessment indicators to comprehensively evaluate a company's quality management system and provide certification results. The user management and interaction module is used to assign user roles, and users can perform data management and evaluation / authentication management within the scope of their assigned permissions.

2. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 1, characterized in that, The data acquisition module includes: Automatic data acquisition interface unit, used for: Connect to various internal management systems of the enterprise through API interfaces or databases, including at least ERP system, CRM system, SCM system, QMS system, R&D system, production system and human resources system; Connect to external data sources via file transfer protocols to obtain data from supplier systems, acquire audit reports from third-party certification bodies via file download, and obtain industry standards and legal requirements from the network via API interfaces; Connect to social media platforms via APIs to obtain tweets, user comments, browsing data, and interaction data about the company's products and services; The data acquisition engine unit is used for: Schedule data acquisition tasks according to the configured timetable or triggering conditions; Write an adapter for each data source. The adapter is responsible for communicating with the data source and extracting the required data. Manual data input unit, used for: Data that cannot be automatically collected is supplemented and improved by manual input.

3. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 2, characterized in that, The data preprocessing module further includes: Data cleaning includes, but is not limited to, removing duplicate data, handling missing values, and correcting erroneous data; Standardize the data, including but not limited to unifying date format, numerical format, text format, unit conversion, field mapping, and data normalization; Perform quality checks, integrity checks, and accuracy checks on the data.

4. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 3, characterized in that, The data fusion module is used for: Merge data from different data sources that have the same records into a single table and use unique identifiers to associate them; Data from different data sources but describing the same entity at different time periods is merged into a single table, and a timestamp field is introduced to mark the validity period of the data, supporting backtracking queries of historical data; It links data from multiple tables to form a complete data view, using JOIN operations to achieve multi-table joins; it supports star and snowflake schema data warehouse architectures, and uses a rule base for custom rules and automated matching, while leveraging natural language processing technology to assist in field name parsing. Connect related data tables to form a complete data system, and use ETL tools to extract, transform and load data; The merged data is output to the designated data storage location for use in quality management system assessment and certification, and the output data format is ensured to meet the requirements of the quality management system assessment and certification system; the storage path and file format are configured to ensure compatibility with downstream analysis tools.

5. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 4, characterized in that, The data analysis module is used for: Using batch insertion and streaming loading mechanisms, data views and data systems are received from the data fusion module, and predefined patterns are used to verify whether the format and structure of the data meet the expected requirements. Use NumPy to calculate descriptive statistics, including at least the mean, standard deviation, and distribution shape, and use the Z-score method to identify outliers, basic characteristics of the data, and potential problems. The ARIMA model is applied to capture long-term trends and periodic fluctuations. Wavelet transform is used to decompose signal components and reveal hidden patterns. Time series invariance analysis is used to extract features from the data and obtain trend changes and interdependencies between multi-source data. Choose the appropriate model based on the complexity and interpretability of the data. Use the random forest model for the classification and grading problem of quality management system, the random forest regression model for the product quality score prediction problem, and the isolation forest model for quality problem identification. The model parameters were optimized using a grid search method, and the tuning process was accelerated using Bayesian optimization. The model was evaluated and certified using accuracy, recall, and F1 score metrics. For changes and reorganizations of initial data, apply either correlation analysis or principal component analysis to select the most influential features and create composite features to form new feature variables; The extracted features are used to train machine learning models to uncover potential patterns and relationships between data, and the model performance is evaluated and optimized through cross-validation.

6. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 5, characterized in that, The quality management system grading model is used for: Based on the relationship between the quality management system and product quality work, the core evaluation module that reflects the effectiveness of the quality management system is determined. Based on the data and information provided by the data analysis module, specific evaluation standards and grading systems are defined, and a grading model of the organization's quality management system is constructed. The quality management system classification model includes five levels: A, B, C, D, and E.

7. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 6, characterized in that, The specific characteristics of the aforementioned quality management system level are as follows: The characteristics of Grade A quality management are: quality management activities are usually ad hoc and lack standardization and documentation; The characteristics of Grade B quality management are: the establishment of basic quality management processes, with initial documentation; The characteristics of Level C quality management are: the quality management process has been standardized and documented, and employees have received training and follow the quality management process; The characteristics of Level D quality management are: standardized quality management processes that are measured and controlled; The characteristics of Level E quality management are: quality management processes are measured and controlled, and continuously optimized through ongoing improvement and innovation.

8. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 7, characterized in that, The workflow of the assessment and certification module includes the following steps: Specific evaluation indicators are defined according to the quality management system grading model. These evaluation indicators include, but are not limited to, the degree of documentation of the quality management process, the status of quality training, the frequency of quality audits, the frequency of quality data collection and analysis, and the mechanisms and records for continuous quality improvement. This involves transforming qualitative evaluation indicators into quantitative evaluation standards; Develop scoring criteria for each evaluation indicator and quantify the score for each indicator; The importance of each indicator in the overall evaluation is determined by the analytic hierarchy process, weights are assigned to each indicator, and the total score is calculated.

9. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 8, characterized in that, The workflow of the assessment and certification module also includes the following steps: The company's quality management level is classified into the corresponding quality management system level based on the total score; The preliminary assessment results were reviewed to confirm the completeness and accuracy of the data. Any problems found were recorded, and suggestions for improvement were put forward. Items that do not meet the requirements will be rejected, marked, and comprehensively evaluated to determine whether they will affect the final evaluation results. Summarize the results of various assessments, conduct a comprehensive analysis, and finally determine the level of the quality management system based on the comprehensive evaluation results; Generate a final evaluation report and provide a certification certificate; Explain the assessment results to the company, propose specific quality management improvement suggestions based on the assessment results, and develop an improvement plan.

10. The quality management system assessment and certification system based on multi-source heterogeneous data processing according to claim 9, characterized in that, The user management and interaction module includes: Enterprise management unit, used for: Please upload additional documents for the quality management assessment. View the assessment progress in real time and understand the current assessment stage; Receive the final evaluation and certification results; View the detailed assessment report and improvement recommendations; The assessment and certification management unit is used for: Configure the data acquisition task, including at least the acquisition frequency, data source, and acquisition rules; Monitor data acquisition status in real time and view data acquisition logs; Receive notifications of anomalies during the data acquisition, data preprocessing, data fusion, data analysis, and evaluation and certification processes; Review and approve the materials submitted by the company; Update the assessment progress and record the completion status of each assessment and certification stage; The user management unit is used for: Supports registration and login for both businesses and assessment / certification bodies; Different permissions are assigned based on user roles.

Images