Data analysis method and system for visually analyzing quality condition of equipment

Abstract

The invention relates to the technical field of equipment quality data analysis, and discloses a data analysis method and system for visually analyzing equipment quality conditions. The method comprises the steps of obtaining an equipment quality index system, and determining data acquisition items according to the system; establishing a multi-source data acquisition platform, and integrating equipment use data, organization quality data and personnel data; performing cleaning and standardization processing on the data; fusing the data by taking the equipment unique identifier as an association key to form a standardized data set; according to different role requirements of an equipment user, a group quality management layer, each enterprise quality management layer and the like, an equipment overall-level analysis model, a system-level analysis model and a product-level analysis model are adaptively called to carry out multi-level mining analysis; results are displayed through a visual chart, and data screening, graph linkage and report export interaction functions are supported. The system is deployed by adopting a B / S architecture. According to the method, multi-source data fusion and multi-dimensional visual analysis are realized, and the efficiency of equipment quality condition analysis and the decision support capability are improved.

Description

Technical Field

[0001] This invention relates to the field of equipment quality data analysis technology, specifically to a data analysis method and system for visually analyzing the quality status of equipment. Background Technology

[0002] In current equipment quality management, equipment quality status analysis and quality decision-making are crucial for ensuring reliable equipment operation and enhancing combat capabilities. With the increasing informatization and complexity of equipment, the data generated throughout its lifecycle exhibits significant characteristics of being massive, multi-source, and heterogeneous. This data includes not only usage data generated during equipment operation, such as runtime, failure frequency, and performance parameters of key components, but also management data accumulated by the organization during quality control, such as quality management system audit records, supply chain quality assessments, and personnel skills certification information. However, existing technologies have significant shortcomings in the collection, analysis, and utilization of these data resources, making it difficult to meet the demands of modern, refined equipment quality management.

[0003] Data silos are a serious problem, making effective integration difficult. Equipment quality data is typically scattered across different departments, business systems, and even the local storage of managers and technicians, such as Product Data Management (PDM) and Manufacturing Execution System (MES). These data vary in format and standard, and contain a large amount of redundancy, missing data, and outliers. Although some research has attempted to apply multi-source information fusion and advanced algorithms, how to efficiently integrate multimodal data from sensors, images, and sound, and accurately classify equipment status levels, remains a pressing technical challenge. Traditional methods struggle to achieve effective correlation and deep integration of cross-system and cross-format data, resulting in numerous "information silos" that hinder a comprehensive and accurate understanding of equipment quality status.

[0004] The analytical methods used are often simplistic and inadequate for in-depth analysis of complex quality conditions. Currently, the analysis of equipment quality conditions largely relies on traditional statistical methods or single algorithmic models, such as simple mean calculation, trend statistics, or k-means clustering. These methods are easily affected by initial conditions and have weak generalization ability when dealing with high-dimensional, nonlinear, and uncertain quality data. Especially in scenarios with imbalanced samples, such as a small number of defect samples, it is difficult to train accurate intelligent identification models. Furthermore, most analytical methods fail to fully utilize the spatiotemporal characteristics and multi-dimensional interactions of equipment operation data, lacking a systematic strategy for multi-level and refined classification of equipment status, resulting in insufficient depth in early warning of potential quality risks and root cause localization of failures.

[0005] The visualization level of analysis results is low, and it is difficult to explicitly express implicit knowledge. The existing presentation methods of analysis results mainly rely on static tables and text reports, lacking intuitive and interactive visualization means. This makes it difficult to intuitively discover and understand the laws and trends hidden behind a large amount of data, such as quality differences between different batches, fault conduction paths, and periodic fluctuations. Although some studies attempt to show data trends through line charts, bar charts, etc., they are weak in supporting multi-dimensional data screening, drilling, and multi-view linkage analysis, limiting the depth and efficiency of data analysis and also affecting the effective transmission and consensus formation of analysis results among decision-makers at all levels.

[0006] It fails to effectively adapt to the different analysis needs of users at different levels. Most of the existing equipment quality analysis tools are general statistical software or dedicated systems with fixed functions, and are not customized for the core demands of different roles such as equipment users (concerned about overall performance), group quality management levels (concerned about process monitoring and resource allocation), and enterprise quality management levels (concerned about root cause analysis and production improvement). Users often need to carry out complex data preprocessing and modeling work, resulting in low analysis efficiency. At the same time, the lack of role-based dynamic permission management and personalized analysis view configuration functions leads to the disconnection between the analysis system and the actual business scenarios of users, and cannot provide accurate, direct, and efficient decision-making support for quality management activities at different levels.

[0007] In summary, the existing technologies have prominent problems in equipment quality data analysis, such as difficult data fusion, shallow analysis dimensions, non-intuitive display, and poor demand adaptation.

[0008] Therefore, there is an urgent need in this field for a data analysis method and system for equipment quality status that can effectively integrate multi-source heterogeneous data, construct multi-dimensional adaptive analysis models, provide intelligent interactive visualization displays, and can accurately adapt to the needs of users at different levels, so as to overcome the limitations of the existing technologies and achieve refined, intelligent, and scientific equipment quality management. Summary of the Invention

[0009] To solve the above deficiencies of the existing technologies, the purpose of the present invention is to provide a data analysis method and system for visually analyzing equipment quality status, which uses multi-source data fusion and multi-dimensional visual analysis to improve the efficiency of equipment quality status analysis and decision-making support capabilities.

[0010] Specifically, On the one hand, the present invention provides a data analysis method for visually analyzing equipment quality status, which includes the following steps: S1, obtaining an index system: obtaining a preset index system for equipment quality status, where the index system covers quality evaluation indexes at the equipment overall, system level, and product level; S2, Planning Data Collection: Determine the specific data collection model based on the indicator system; based on this, clarify the data collection items for equipment use, organizational quality, and organizational personnel. S3, Establish a multi-source data acquisition platform: Based on the data acquisition model determined in S2, establish a military equipment data acquisition platform, integrate multiple data sources, and collect equipment usage data, organizational quality data, and organizational personnel data; S4. Data cleaning and multi-dimensional data verification: Perform data cleaning and verification on the raw data obtained from the data acquisition platform to ensure data quality. S5, Data Standardization: Standardize the cleaned data; convert data of different formats but representing the same semantics into a system-defined standard format. S6, Data Fusion and Storage: The standardized equipment usage data, organizational quality data, and organizational personnel data are linked to form a complete and standardized comprehensive equipment quality dataset, which is then stored in the database. S7, Establish a multi-level analysis model: Adaptively call the corresponding analysis model according to different user roles and their analysis needs, and perform mining analysis on the fused standard dataset; S8, Role-Based Access Control: Configures data access and operation permissions based on role permissions, allowing for fine-grained permission configuration according to different personnel's job level, job responsibilities, and data sensitivity; S9. Visualization and Intelligent Interaction: Displays analysis results through visual charts and provides functions such as data filtering and drill-down, multi-view linkage analysis, and analysis report generation and export.

[0011] Furthermore, the data cleaning in step S3 includes: identifying and removing completely duplicate data records, marking and correcting outliers that are clearly outside the reasonable range; filling in missing values ​​according to business rules; and the multi-dimensional data verification in step S3 includes: cross-validating logical consistency through multi-dimensional query statistics.

[0012] Furthermore, the standardization process in step S4 includes: unifying the date and time data into the YYYY-MM-DD format; performing a unified conversion of the unit system; and using standardized encoding for the text classification information.

[0013] Furthermore, the multi-level analysis model in step S6 includes: an overall equipment analysis model, a product-level analysis model, and a system-level analysis model. The overall equipment analysis model is geared towards equipment users. By calculating the mean time between failures (MFHBF) and macro-level indicators of mission reliability, and performing trend analysis on these indicators, it achieves a predictive assessment of the overall health status of the equipment, providing macro-level data support for equipment operation and maintenance decisions. The product-level analysis model is geared towards the group's quality management level. It identifies differences by horizontally comparing similar quality indicators across different aircraft models and military units, and vertically analyzes the trend changes of key parameters in specific systems, such as power systems and avionics systems, to detect abnormal fluctuations. It can also introduce process capability indices, such as Cp and Cpk, to assess the stability of key processes. Finally, it generates a comprehensive quality analysis report to assist management in resource allocation and process optimization decisions. The system-level analysis model is geared towards the quality management levels of individual enterprises. It identifies major failure modes by implementing Failure Mode and Effects Analysis (FMEA), and uses correlation analysis to explore the relationship between product failures and multiple factors such as production batches and operating environment, analyzing and locating the root causes of product failures.

[0014] Furthermore, the visualization chart in step S9 is dynamically generated by the chart generation engine; wherein, The chart generation engine adaptively matches and renders the corresponding chart from a predefined set of chart types based on the number of dimensions, measurement type, and user selection of the data to be analyzed. The set of chart types includes at least line charts, bar charts, pie charts, and scatter plots. Data filtering and drill-down: Provides users with a wealth of filtering conditions, including but not limited to time range, aircraft type, specific location, cause of failure, etc. The time range referred to here is such as: the past month, the current quarter, the current year, etc. By combining these conditions, users can quickly focus on the subset of data they are interested in and drill down to obtain more detailed information. Multi-view linkage analysis: Supports intelligent linkage between different chart views; Analysis report generation and export: Capture all visual chart views and effective data filtering statuses in the current step S8; assemble the chart views and filtering statuses into a structured document according to the preset report template; output the structured document in PDF or Word format.

[0015] On the other hand, the present invention also provides an analysis system for a data analysis method for visually analyzing equipment quality status, comprising: a data source integration module, a multi-dimensional analysis model construction module, a user requirement adaptation module, a visualization graph generation module, an interactive operation module, and a system management and support module. The data source integration module is responsible for the access, preprocessing and fusion of multi-source heterogeneous data, which includes equipment usage data acquisition unit, organizational quality data acquisition unit, data cleaning unit, data standardization unit and data fusion unit; The multi-dimensional analysis model building module is the system's analysis engine, which includes product-level analysis units, system-level analysis units, and equipment-wide analysis units. The user requirement adaptation module binds the system's analytical capabilities to user roles, enabling on-demand analytical services. It includes equipment user requirement configuration units, group quality management system requirement configuration units, and enterprise quality management system requirement configuration units. The visualization graphics generation module is based on modern Web visualization technology and includes bar chart generation unit, line chart generation unit, pie chart generation unit, and other graphics generation units; The interactive operation module provides users with an intuitive and easy-to-use human-computer interaction interface, which includes a data filtering unit, a graphical linkage unit, and a report export unit. The system management and support module provides basic public services such as unified user authentication, access control, system configuration, operation monitoring, data storage, and external interfaces. It includes a system deployment unit, an external system integration unit, an access control unit, and a data storage and management unit.

[0016] Furthermore, the external system interface unit of the system management and support module provides standard APIs for data exchange with internal enterprise systems.

[0017] Furthermore, the interactive operation module's graphical linkage unit supports multi-view event response, automatically updating the data display of related views when a user clicks on a chart element.

[0018] Furthermore, the system deployment unit adopts a B / S architecture and is deployed on the enterprise's intranet cloud server, supporting multi-terminal access via browser and synchronizing data with external data warehouses through HTTP interface.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: I. This invention integrates multi-dimensional data such as equipment usage and organizational quality to construct a three-level analysis model at the product level, system level, and overall equipment level. This breaks through the limitations of traditional analysis with a single dimension and can comprehensively and accurately characterize the equipment quality status at different levels.

[0020] Second, this invention utilizes a variety of interactive charts to display the analysis results, transforming implicit patterns such as fault trends and batch differences that are difficult to find in traditional reports into intuitive graphics. Combined with drill-down and linked analysis, it significantly improves the readability and insight efficiency of the results.

[0021] Third, this invention has customized different analysis dimensions and indicators to meet the different needs of equipment users, group and enterprise quality management departments, etc. Users can quickly obtain the required conclusions without complicated data processing, which greatly improves the pertinence and efficiency of quality data analysis.

[0022] Fourth, by providing information such as root cause analysis and trend prediction, this invention provides direct data support for management to formulate improvement strategies and optimize processes, promotes the transformation of quality management from post-event handling to pre-event early warning, and helps group enterprises reduce quality costs and improve equipment quality levels. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the data analysis method for visually analyzing equipment quality status according to the present invention. Figure 2 This is an overall architecture diagram of the data analysis system for visually analyzing the quality status of equipment according to the present invention. Detailed Implementation

[0024] This invention provides a data analysis method for visually analyzing the quality status of equipment. This method systematically collects, processes, integrates, and analyzes multi-source data related to equipment quality, and then visualizes the data based on different user needs, achieving multi-level and multi-dimensional insights into the equipment's quality status. The preferred implementation of this method is described in detail below using a certain type of aircraft as an example. It is understood that the application of this method is not limited to this, and can also be applied to other types of military equipment or complex equipment. Figure 1 As shown, it includes the following steps: S1. Obtain the indicator system; Obtain a pre-defined equipment quality status indicator system. This system is typically hierarchical, covering quality evaluation indicators at different levels, such as overall equipment, system level, and product level. These indicators include macro-level metrics like reliability, maintainability, testability, and safety, as well as specific performance parameter thresholds. This indicator system provides the fundamental basis for subsequent data collection and analysis.

[0025] S2, Planning Data Acquisition; The specific data collection model is determined based on the indicator system. Based on this, the data collection items for equipment use, organizational quality, and organizational personnel are defined. As a preferred implementation method in this embodiment, these data collection items specifically include: The specific data collection items for equipment use are: equipment model, unique number, product composition information, field-reported fault data, usage quality data, cumulative service time, and current condition; among which, field-reported fault data includes fault code, fault phenomenon, time, location, performance parameter deviation value, and mission completion rate.

[0026] The specific items for collecting organizational quality data include: quality audit cycle, quality management system effectiveness evaluation score, process compliance inspection records, and supplier qualification assessment reports.

[0027] The specific data collection items for personnel include: basic identity information of personnel, department, job level, operating permissions, training records and skills certification status.

[0028] S3. Establish a multi-source data acquisition platform; Based on the data acquisition model determined in S2, a military equipment data acquisition platform is established to integrate multiple data sources and collect equipment usage data, organizational quality data, and organizational personnel data. As a preferred embodiment of this practice, the platform specifically includes: Equipment usage data collection: This is used to collect data generated during actual use of equipment, such as equipment model, unique number, product composition information, field-reported fault data, usage quality data, cumulative service time, and current condition. Among them, field-reported fault data includes fault codes, fault phenomena, time, location, performance parameter deviation values, and mission completion rates.

[0029] Organizational quality data collection: This is used to collect data related to the operation of the quality management system, such as quality audit cycles, quality management system effectiveness evaluation scores, process compliance inspection records, and supplier qualification assessment reports.

[0030] Organizational personnel data collection: This is used to collect personnel information related to equipment quality assurance, such as basic personnel identity information, department, job level, operating authority, training records, and skills certification status.

[0031] The platform can integrate data from various data sources, such as sensors and existing business management systems like PDM and MES, through data interfaces, ensuring the breadth and timeliness of the data.

[0032] S4. Data cleaning and multi-dimensional verification; The raw data acquired by the data acquisition platform undergoes data cleaning and verification to ensure data quality. This process includes: Data cleaning: Perform multiple rounds of data cleaning to identify and remove completely duplicate data records; mark and correct outliers that are clearly outside the reasonable range, such as unreasonable maximum or minimum values ​​or data that do not conform to physical laws; fill in missing values ​​according to business rules, which can be done or explained according to business rules.

[0033] Multi-dimensional data verification: After cleaning, logical consistency is cross-validated through multi-dimensional queries and statistics. By querying and statistically analyzing multiple dimensions, such as by time, by equipment model, and by fault type, the integrity and logical consistency of the data are cross-validated to ensure data accuracy and usability.

[0034] S5, Data Standardization; To address the issue of heterogeneous formats in multi-source data, the cleaned data underwent standardization. Data in different formats but representing the same semantics were uniformly converted to a system-defined standardized format. For example, date and time data in various formats were unified into the standard format YYYY-MM-DD, including formats such as 20231001, 2023 / 10 / 01, and 01-OCT-2023. Different unit systems, such as pressure units MPa and psi, were standardized for conversion. Text classification information was encoded using standardized methods. This approach aims to eliminate data ambiguity and lay the foundation for subsequent fusion analysis.

[0035] S6, Data Fusion and Storage; The standardized equipment usage data, organizational quality data, and organizational personnel data are linked to form a complete and standardized comprehensive equipment quality dataset, which is then stored in a database. In a preferred embodiment, the unique identifier of the equipment, such as the aircraft serial number or product serial number, is used as the key to link data from the equipment usage data collection unit with corresponding organizational quality data and relevant personnel, forming a complete and standardized comprehensive equipment quality dataset. In another preferred embodiment, this dataset is ultimately persistently stored in the system's relational database for efficient querying and analysis.

[0036] S7. Establish a multi-level analysis model; The system adaptively calls the corresponding analysis model based on different user roles and their analysis needs to perform mining analysis on the fused standard dataset. As a preferred implementation method in this embodiment, the specific implementation is as follows: a) Overall Equipment Analysis Model: This model is designed for equipment users and assesses overall performance. This level focuses on the overall quality risk and availability of the equipment. By calculating macro-level indicators such as Mean Time Between Failures (MFHBF) and mission reliability, and performing trend analysis on these indicators, it achieves a predictive assessment of the overall health status of the equipment, providing macro-level data support for equipment operation and maintenance decisions.

[0037] b) Product-level analysis model: This model is targeted at the group's quality management level for process monitoring and analysis. This level focuses on identifying quality risk points and weak links in work at each system level of the equipment. Specifically, it includes: horizontally comparing similar quality indicators of different models and military units to identify differences; vertically analyzing the trend changes of key parameters of specific systems, such as power systems and avionics systems, to detect abnormal fluctuations; and introducing process capability indices, such as Cp and Cpk, to evaluate the stability of key processes. Finally, a comprehensive quality analysis report is generated to assist the management in resource allocation and process optimization decisions.

[0038] c) System-level analysis model: This model is targeted at the quality management levels of each enterprise for root cause analysis and location of products. Specifically, this level focuses on the deviation of specific products (components / sub-systems) from the quality baseline, and identifies the main failure modes by implementing Failure Mode and Effects Analysis (FMEA); uses correlation analysis to explore the associations between product failures and multiple factors such as production batches and usage environments, and analyzes and locates the root causes of products. Based on this, a problem cause analysis report is generated to accurately guide production process improvement or on-site maintenance.

[0039] S8. Role-based permission management; Configure data access and operation permissions based on roles. According to the ranks, job responsibilities of different personnel, and data sensitivity, perform refined permission configuration. Permission management not only limits the types and ranges of data that users can access (for example, only view data of specific models or specific time periods), but also controls their specific operation functions, such as only allowing viewing, allowing data export, allowing configuration of analysis models, etc., so as to ensure data security and compliance while ensuring sufficient data sharing.

[0040] S9. Visualization display and intelligent interaction; Display the analysis results through visual charts, and provide functions such as data screening and drilling, multi-view linked analysis, and generation and export of analysis reports. The visual chart generation engine generates charts dynamically. As a preferred implementation manner of this embodiment, the specific implementation is as follows: Generate visual charts: Automatically convert the analysis results in S7 into various visual charts. Adaptively match and render the corresponding charts from the predefined chart type set. At the same time, provide line charts (for showing trends), bar charts (suitable for comparison), pie charts or sector charts (suitable for showing composition ratios), scatter plots, etc. for users to select as needed. Preferably, the charts clearly label the coordinate axes, legends, and data points.

[0041] Data filtering and drill-down: Provides users with a wide range of filtering options, including but not limited to time range, aircraft type, specific location, and cause of failure. Time ranges include, for example, the past month, the current quarter, and the current year. By combining these options, users can quickly focus on the subset of data they are interested in and drill down to obtain more detailed information.

[0042] Multi-view linked analysis: Supports intelligent linkage between different chart views. For example, when a user views a bar chart comparing the number of failures of different products, clicking on a product series in the chart will automatically redirect or update the interface after verifying permissions, displaying a pie chart of the distribution of quality analysis causes for that product or other detailed information views, thereby achieving seamless traceability from macro overview to micro details and deepening the dimensions of analysis.

[0043] Analysis report generation and export: Capture all visual chart views and effective data filtering statuses in the current step S8; assemble the chart views and filtering statuses into a structured document according to the preset report template; output the structured document in PDF or Word format.

[0044] See Figure 2 This invention also provides an analysis system for a data analysis method that visualizes and analyzes the quality status of equipment, implementing the above-described method. This system adopts a layered and modular design concept, possessing good scalability and integration capabilities. A preferred embodiment of the system includes the following modules: 1. Data Source Integration Module 1: This module is the foundation of the system and is responsible for the access, preprocessing, and fusion of multi-source heterogeneous data. This module includes Equipment Usage Data Acquisition Unit 11, Organizational Quality Data Acquisition Unit 12, Data Cleaning Unit 13, Data Standardization Unit 14, and Data Fusion Unit 15.

[0045] 2. Multi-dimensional Analysis Model Construction Module 2: This module is the system's analysis engine, preferably with built-in or supporting the integration of multiple analysis models. This module includes a product-level analysis unit 21, a system-level analysis unit 22, and an overall equipment-level analysis unit 23. These units encapsulate the corresponding analysis algorithms and logic, which can be called on demand by the user-requirement adaptation module. Preferably, the product-level analysis unit of the multi-dimensional analysis model construction module integrates a fault prediction and health management (PHM) model, and the system-level analysis unit supports statistical process control (SPC) analysis.

[0046] 3. User Needs Adaptation Module 3: This module binds the system's analytical capabilities to user roles, enabling on-demand analytical services. This module includes equipment user needs configuration unit 31, group quality management system needs configuration unit 32, and individual enterprise quality management system needs configuration unit 33. By pre-setting or customizing analytical views, indicators, and report templates for different roles, users at different levels can obtain analytical content that matches their responsibilities.

[0047] 4. Visualization Module 4: This module, based on modern Web visualization technology, provides rich, customizable, and high-performance chart generation capabilities. It preferably includes a bar chart generation unit 41, a line chart generation unit 43, a pie chart generation unit 42, and other chart generation units 44, capable of responding to front-end interactive commands and dynamically generating and updating charts. These other chart generation units include, for example, scatter plots and radar charts.

[0048] 5. Interactive Operation Module 5: This module provides users with an intuitive and easy-to-use human-computer interaction interface. Preferably, it includes a data filtering unit 51 providing filtering controls and logic, a graph linkage unit 52, and a report export unit 53. The graph linkage unit 52 is responsible for handling the linkage events and state synchronization between charts. Preferably, the graph linkage unit 52 supports multi-view event response; when a user clicks on a chart element, the data display of the associated views is automatically updated. The report export unit 53 is responsible for assembling the analysis results into a report file for download.

[0049] 6. System Management and Support Module 6: This module ensures the stable, secure, and efficient operation of the entire system by providing unified user authentication, access control, system configuration, operation monitoring, data storage, and external interfaces. Preferably, it includes a system deployment unit 61, an external system interface unit 62, an access control unit 63, and a data storage and management unit 64.

[0050] System Deployment Unit 61: Deployed on the enterprise intranet cloud server using a B / S architecture, it supports multi-terminal access via browser and synchronizes data with external data warehouses through an HTTP interface. Users do not need to install a dedicated client; they can access the system simply by entering the URL in a browser and logging in. It supports various terminals such as PCs and tablets.

[0051] External system interface unit 62 is used to provide standard APIs for data exchange with internal enterprise systems.

[0052] Permission Management Unit 63: Responsible for managing user accounts, role definitions, and permission allocation, supporting S8's permission control.

[0053] Data storage and management unit 64: Responsible for persistent storage and management of standardized equipment quality datasets, system configurations, user information, logs, etc.

[0054] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A data analysis method for visually analyzing the quality status of equipment, characterized in that, It includes the following steps: S1, Obtain the indicator system: Obtain the preset equipment quality status indicator system, which covers the overall equipment, system level and product level quality evaluation indicators; S2, Planning Data Collection: Determine the specific data collection model based on the indicator system; based on this, clarify the data collection items for equipment use, organizational quality, and organizational personnel. S3, Establish a multi-source data acquisition platform: Based on the data acquisition model determined in S2, establish a military equipment data acquisition platform, integrate multiple data sources, and collect equipment usage data, organizational quality data, and organizational personnel data; S4. Data cleaning and multi-dimensional data verification: Perform data cleaning and verification on the raw data obtained from the data acquisition platform to ensure data quality. S5, Data Standardization: Standardize the cleaned data; convert data of different formats but representing the same semantics into a system-defined standard format. S6, Data Fusion and Storage: The standardized equipment usage data, organizational quality data, and organizational personnel data are linked to form a complete and standardized comprehensive equipment quality dataset, which is then stored in the database. S7, Establish a multi-level analysis model: Adaptively call the corresponding analysis model according to different user roles and their analysis needs, and perform mining analysis on the fused standard dataset; S8, Role-Based Access Control: Configures data access and operation permissions based on role permissions, allowing for fine-grained permission configuration according to different personnel's job level, job responsibilities, and data sensitivity; S9. Visualization and Intelligent Interaction: Displays analysis results through visual charts and provides functions such as data filtering and drill-down, multi-view linkage analysis, and analysis report generation and export.

2. The data analysis method for visually analyzing equipment quality status according to claim 1, characterized in that, Data cleaning in step S3 includes: identifying and removing completely duplicate data records, marking and correcting outliers that are clearly outside the reasonable range; filling in missing values ​​according to business rules; multi-dimensional data verification in step S3 includes: cross-validating logical consistency through multi-dimensional query statistics.

3. The data analysis method for visually analyzing equipment quality status according to claim 1, characterized in that, The standardization process in step S4 includes: unifying date and time data into YYYY-MM-DD format; performing standardized conversion of unit systems; and using standardized encoding for text classification information.

4. The data analysis method for visually analyzing equipment quality status according to claim 1, characterized in that, Step S6 involves a multi-level analysis model, including: an overall equipment analysis model, a product-level analysis model, and a system-level analysis model. The overall equipment analysis model is designed for equipment users. It calculates the mean time between failures (MFHBF) and macro-level indicators of mission reliability, and performs trend analysis on these indicators to provide predictive assessments of the overall equipment health status, offering macro-level data support for equipment operation and maintenance decisions. The product-level analysis model is designed for the group's quality management team. It identifies differences by comparing similar quality indicators across different aircraft models and military units, and analyzes the trend changes of key parameters in specific systems, such as power systems and avionics systems, to detect abnormal fluctuations. It can also introduce process capability indices, such as Cp and Cpk, to assess the stability of key processes. Finally, it generates a comprehensive quality analysis report to assist management in resource allocation and process optimization decisions. The system-level analysis model is designed for the quality management teams of individual enterprises. It identifies major failure modes by implementing Failure Mode and Effects Analysis (FMEA), and uses correlation analysis to explore the relationship between product failures and multiple factors such as production batches and operating environment, analyzing and locating the root causes of product failures.

5. The data analysis method for visually analyzing equipment quality status according to claim 1, characterized in that, The visualization chart in step S9 is dynamically generated by the chart generation engine; wherein, The chart generation engine adaptively matches and renders the corresponding chart from a predefined set of chart types based on the number of dimensions, measurement type, and user selection of the data to be analyzed. The set of chart types includes at least line charts, bar charts, pie charts, and scatter plots. Data filtering and drill-down: Provides users with a wealth of filtering conditions, including but not limited to time range, aircraft type, specific location, cause of failure, etc. The time range referred to here is such as: the past month, the current quarter, the current year, etc. By combining these conditions, users can quickly focus on the subset of data they are interested in and drill down to obtain more detailed information. Multi-view linkage analysis: Supports intelligent linkage between different chart views; Analysis report generation and export: Capture all visual chart views and effective data filtering statuses in the current step S8; assemble the chart views and filtering statuses into a structured document according to the preset report template; output the structured document in PDF or Word format.

6. An analysis system for a data analysis method for visually analyzing the quality status of equipment according to any one of claims 1-5, characterized in that, It includes: a data source integration module, a multi-dimensional analysis model construction module, a user requirement adaptation module, a visualization graph generation module, an interactive operation module, and a system management and support module. The data source integration module is responsible for the access, preprocessing and fusion of multi-source heterogeneous data, which includes equipment usage data acquisition unit, organizational quality data acquisition unit, data cleaning unit, data standardization unit and data fusion unit; The multi-dimensional analysis model building module is the system's analysis engine, which includes product-level analysis units, system-level analysis units, and equipment-wide analysis units. The user requirement adaptation module binds the system's analytical capabilities to user roles, enabling on-demand analytical services. It includes equipment user requirement configuration units, group quality management system requirement configuration units, and enterprise quality management system requirement configuration units. The visualization graphics generation module is based on modern Web visualization technology and includes bar chart generation unit, line chart generation unit, pie chart generation unit, and other graphics generation units; The interactive operation module provides users with an intuitive and easy-to-use human-computer interaction interface, which includes a data filtering unit, a graphic linkage unit, and a report export unit. The system management and support module provides basic public services such as unified user authentication, access control, system configuration, operation monitoring, data storage, and external interfaces. It includes a system deployment unit, an external system integration unit, an access control unit, and a data storage and management unit.

7. The analysis system of the data analysis method for visual analysis of equipment quality status according to claim 6, characterized in that, The system management and support module is an external system interface unit that provides standard APIs for data exchange with internal enterprise systems.

8. The analysis system of the data analysis method for visual analysis of equipment quality status according to claim 6, characterized in that, The interactive operation module's graphical linkage unit supports multi-view event response. When a user clicks on a chart element, the data display of the associated views is automatically updated.

9. The analysis system of the data analysis method for visual analysis of equipment quality status according to claim 6, characterized in that, The system deployment unit adopts a B / S architecture and is deployed on the enterprise's intranet cloud server. It supports multi-terminal access via browser and synchronizes data with external data warehouses through HTTP interface.

Images