Data governance methods and systems

Abstract

The present invention relates to the field of data processing technology, and provides a data governance method and system, which obtains stock data in at least one data module, filters out key data in the stock data; performs standardization processing on the key data, and determines data standards; generates rules based on the data standards, verifies the stock data and / or standardizes the newly added data in the data module based on the rules, uses the verified stock data and / or the standardized newly added data as standard data, and stores the standardized newly added data in the corresponding data module. Through the implementation of this method and the application of the system, standards are formulated according to the stock data in each data module, and then rules are generated according to the standards. Finally, the stock data in the data module and the newly added data to be stored in the data module are standardized based on the rules, which unifies the interfaces and data formats of each data module, reduces the difficulty of data integration, improves the usability of data, and realizes closed-loop governance of data.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a data governance method and system. Background Art

[0002] Currently, the digital economy has become a key engine driving high-quality economic development. As a core asset for businesses, fully leveraging data to drive business innovation and development has become a consensus across all industries. In the digital transformation of traditional manufacturing companies, data has become a key production factor driving the digitalization, networking, and intelligent development of enterprises. However, manufacturing processes are long, and data runs through every link, including product R&D and design, production and manufacturing, process flow, procurement, supply and marketing, logistics, after-sales service, business management, and operation and maintenance services. The amount of data in each system is huge and complex. How to resolve data silos, improve enterprise data quality, unify data standards, and unlock data value, thereby allowing data to better serve and empower businesses, is a major challenge facing the digitalization of manufacturing enterprises.

[0003] like Figure 1 As shown in FIG, it is the structure of an existing manufacturing data system, which includes multiple data modules distributed among different subsystems, such as a product data management (PDM) data module 41, an enterprise resource planning (ERP) data module 42, a manufacturing execution system (MES) data module 43, a maintenance, repair & operations (MRO) data module 44, a quality management system (QMS) data module 45, a warehouse management system (WMS) data module 46, etc. These data modules can store different data, and different data modules are connected to each other. The data formats in the data modules are messy and the data integration is difficult, which makes it difficult to standardize the stock data in each data module and to constrain the newly added data to make it comply with the standards, resulting in poor data usability. Summary of the Invention

[0004] The present invention provides a data governance method and system to solve the defects in the prior art of data systems, such as the messy data formats of various data modules, the high difficulty in data integration, the inconsistent timeliness and accuracy of data maintenance, which make it difficult to standardize the existing data in each data module and difficult to constrain the newly added data to make it comply with the standards. The method and system can standardize the existing data in each module and constrain the newly added data to make it comply with the standards.

[0005] The present invention provides a data governance method, comprising:

[0006] Obtaining stock data in at least one data module;

[0007] Filtering out key data from the stock data;

[0008] Standardize the key data and determine data standards;

[0009] generating rules based on the data standards;

[0010] The existing data and / or the newly added data are verified based on the rules, and the verified existing data and / or the newly added data after standardization are used as standard data.

[0011] According to a data governance method provided by the present invention, key data in the stock data is screened out, including:

[0012] Calculating the importance of each data in the stock data;

[0013] sorting the data based on the importance;

[0014] Select a preset number of data in the top as key data.

[0015] According to a data governance method provided by the present invention, calculating the importance of each data in the stock data includes:

[0016] Calculate the initial citation degree of each data in the stock data:

[0017] (αCI n +(1-α)*(DI n +SA x )),

[0018] Where α is the damping coefficient, CI n is the citation degree of the nth data, DI n is the completeness of the nth data, SA x is the stability under x period, x is the time period;

[0019] Superimpose citations based on the data's citation status:

[0020]

[0021] Among them, p is the total number of data; Out n is the number of outlinks of the nth data reference node;

[0022] Calculate the importance of data:

[0023]

[0024] Among them, TR is importance;

[0025] Iterative calculation is performed until the importance tends to be stable, and the final importance result is obtained.

[0026] According to a data governance method provided by the present invention, the calculation of the citation degree includes:

[0027] CI=(SQ n +50*US n )

[0028] Among them, SQ n Number of queries served for the nth service; US n The number of downstream users of the nth service;

[0029] The stability calculation includes:

[0030] Calculation statistical period:

[0031] Cycle h =(30*24-En h ) / (30*24)

[0032] Cycle d =(30-En d ) / 30

[0033] Cycle w =(7-En w ) / 7

[0034] Among them, Cycle h Cycle is the statistical cycle in hours. d Cycle is the statistical cycle with days as the time unit. w is the statistical period with weeks as the time unit, En h The number of alarms that are not generated on time within a time period in hours, En d The number of alarms that were not generated on time within a time period of days, En wThe number of alarms that were not generated on time within a time period of weeks;

[0035] Computational stability:

[0036] SA x =(Cycle x *100),

[0037] Where x is the time period.

[0038] According to a data governance method provided by the present invention, the key data is standardized to determine the data standard, including:

[0039] Distinguishing the attribute information of the key data and deleting the attribute information that meets the preset redundant information conditions;

[0040] Determine characteristic attributes of each attribute information in the key data, and construct a characteristic attribute set of each attribute information;

[0041] The characteristic attribute sets of the various attribute information are combined to construct a data standard set.

[0042] According to a data governance method provided by the present invention, generating rules based on the data standards includes:

[0043] Based on the data standard set, a rule including a data format is generated; the data format includes a data definition, a data structure, and a value list.

[0044] According to a data governance method provided by the present invention, verifying the existing data and / or standardizing the newly added data based on the rules includes:

[0045] Verifying the stock data in the data module one by one according to the rules, screening out the stock data whose data formats comply with the rules as standard data, correcting the screened stock data whose data formats do not comply with the rules according to the rules, and using the corrected stock data as standard data;

[0046] And / or, the newly added data is verified according to the rule, and the newly added data whose data format complies with the rule is screened out as standard data.

[0047] According to a data governance method provided by the present invention, after the verified existing data and / or the standardized newly added data are used as standard data, the method further includes:

[0048] Panoramic data is constructed based on the association relationship between the standard data and the data counterparts.

[0049] The present invention also provides a data governance system for executing any of the above-mentioned data governance methods, the data governance system comprising a data governance unit and at least one data module, wherein the data module contains stock data; the data governance unit is configured to obtain the stock data in the at least one data module, filter out key data in the stock data, perform standardization on the key data, determine data standards, generate rules based on the data standards, verify the stock data based on the rules, and standardize newly added data based on the rules, and use the verified stock data and the standardized newly added data as standard data;

[0050] The data governance unit includes a basic data support layer, and the basic data support layer includes a standard unit and a rule unit; the standard unit contains the data standard; and the rule unit contains the rule.

[0051] According to a data governance system provided by the present invention, the data governance unit further includes a panoramic data construction layer, and the panoramic data construction layer is used to construct panoramic data based on the association relationship between the standard data and the data counterparts.

[0052] The data governance method and system provided by the present invention obtain the stock data in at least one data module, filter out the key data in the stock data; standardize the key data and determine the data standard; generate rules based on the data standard, verify the stock data and / or standardize the newly added data in the data module based on the rules, and use the verified stock data and / or standardized newly added data as standard data. Through the implementation of this method and the application of the system, standards are formulated according to the stock data in each data module, and then rules are generated according to the standards. Finally, the stock data in the data module and the newly added data to be stored in the data module are standardized based on the rules, thereby unifying the data format of the data in each data module, reducing the difficulty of data integration, improving the usability of data, and realizing closed-loop data governance. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 It is a structural diagram of a data system in the prior art;

[0055] Figure 2 It is a flowchart of the data governance method provided by the present invention;

[0056] Figure 3 is a flowchart of an optional data governance method in an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of an optional data governance system provided by the present invention;

[0058] Figure 5 This is one of the structural diagrams of the data governance unit in an embodiment of the present invention;

[0059] Figure 6 This is the second structural diagram of the data governance unit in an embodiment of the present invention.

[0060] Reference numerals:

[0061] 40: Data governance unit; 41: Product data management data module; 42: Enterprise resource planning data module; 43: Production execution system data module; 44: Operation and maintenance data module; 45: Quality management system data module; 46: Warehouse management system data module; 50: Basic data support layer; 51: Standard unit; 52: Rule unit; 60: Panoramic data construction layer; 61: Business combing module; 62: Data identification module; 63: Data classification and grading module; 64: Table structure module; 65: Data association module. DETAILED DESCRIPTION

[0062] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0063] The following combination Figure 2 and Figure 3 The data governance method of the present invention is described.

[0064] like Figure 1 As shown, the present invention provides a data governance method, comprising the following steps:

[0065] S1. Obtain stock data in at least one data module.

[0066] In this step, it is necessary to obtain stock data from the data module. It should be understood that the present invention does not limit the specific fields, distribution characteristics, quantity and types of data modules. The following is an example of a data system of a manufacturing enterprise:

[0067] The data system of the manufacturing industry includes data modules distributed in different subsystems (such as production execution systems, quality management systems, etc.). These data modules may include product data management (PDM) data modules, enterprise resource planning (ERP) data modules, manufacturing execution system (MES) data modules, maintenance, repair & operations (MRO) data modules, quality management system (QMS) data modules, warehouse management system (WMS) data modules, etc. It should be noted that the above data modules are only examples. In addition to the above data modules, the present invention may also include other related data modules, which can store different data.

[0068] The Product Data Management module manages all product-related information (including parts information, configurations, documentation, CAD files, structures, permissions, etc.) and all product-related processes (including process definition and management). The Enterprise Resource Planning module effectively shares and utilizes enterprise resources. By fully organizing and efficiently transmitting information through information systems, the enterprise's resources are rationally allocated and utilized across all aspects of procurement, storage, production, sales, personnel, finance, and materials, thereby improving operational efficiency. The Production Execution System module forms a comprehensive production information system, meeting the requirements of enterprises at different planning stages in the field of information-based production management and ensuring a smooth transition and gradual improvement in the informationization process. The Operations and Maintenance module utilizes the internet to order consumables (stationery, office supplies, equipment and machine repair parts, etc.) from the enterprise, enabling efficient delivery from routine procurement. The Quality Management System module selects and combines several system elements tailored to the enterprise's specific characteristics to strengthen quality management activities throughout the entire process, from design and development, production, inspection, sales, and use. These elements are then institutionalized and standardized, becoming internal quality management requirements and procedures. The warehouse management system data module is used to manage information, resources, behaviors, inventory and distribution operations in accordance with business rules and operating principles to improve efficiency.

[0069] For manufacturing companies, the aforementioned modules typically contain large amounts of data, known as stock data. This data is crucial to the enterprise and forms the cornerstone of its normal operation. In addition to stock data, new data is generated as production progresses, such as newly purchased equipment, new suppliers, and new standards. This new data, known as incremental data, is an indispensable component of enterprise development. In existing technologies, stock data often exists in a variety of formats. For example, different data modules utilize different data formats, which can be numeric, character, or binary. This creates difficulties in accessing, organizing, sharing, and applying data across different modules. Accessibility is primarily due to the fact that the same data is maintained across multiple systems, resulting in inconsistent timeliness and accuracy, making it difficult to identify an accurate and authoritative data source. Organizational difficulties arise from the fact that global equipment data is dispersed across multiple business systems, increasing the difficulty and cost of data integration. Sharing is primarily due to the numerous interfaces created by data sharing between systems. When the same data is maintained across multiple systems, these interfaces must also be maintained simultaneously, making rapid sharing difficult. Application difficulties arise from the fact that data accuracy cannot be guaranteed due to the different data sources across these interfaces. For existing data, even data in the same data module may exist in multiple different formats, which further increases the difficulty of data sharing and application.

[0070] S2. Filter out key data from the existing data.

[0071] In this step, the importance of each piece of data in the existing data is calculated, ranked based on importance, and a preset number of data items with the highest rankings are selected as key data. Specifically, the importance of the existing data is sorted and screened by leveraging the flow relationships between data items to determine the key data to be standardized. If a piece of data is applied many times, it indicates that it is relatively important, meaning that its TR value is relatively high. If a piece of data with a high TR value is applied to another piece of data, the TR value of the other piece of data will be correspondingly increased.

[0072] The algorithm for calculating the importance of existing data in existing data includes three parts: citation index (CI), data integrity (DI) and stability score (SA). The specific meaning and calculation formula are as follows:

[0073] (1) Calculation of citations

[0074] The citation score is divided into two parts: the score itself and the value transferred. The specific calculation formula is as follows:

[0075] CI=(SQ n +50*US n )

[0076] Among them, SQ n Number of queries served for the nth service; US n The number of downstream users of the nth service.

[0077] (2) Definition of data integrity

[0078] Data integrity represents the degree of completeness of the data. In an optional embodiment of the present invention, the total score of data integrity is 100 points, and the score composition is shown in Table 1.

[0079] Table 1 Data completeness score composition table

[0080] describe Score Data Source 25 points Data Description 20 points Data Annotation 25 points Table Chinese name 15 points Data Hierarchy 10 points Business Theme 10 points

[0081] (3) Stability score

[0082] Calculate the statistical period. In an optional embodiment of the present invention, the statistical period includes hours, days, and weeks. The calculations for hourly scheduling, daily scheduling, and weekly scheduling are as follows:

[0083] Cycle h =(30*24-En h ) / (30*24)

[0084] Cycle d =(30-En d ) / 30

[0085] Cycle w =(7-En w ) / 7

[0086] Among them, Cycle h Cycle is the statistical cycle in hours. d Cycle is the statistical cycle with days as the time unit. w is the statistical period with weeks as the time unit, En h The number of alarms that are not generated on time within a time period in hours, En d The number of alarms that were not generated on time within a time period of days, En w The number of alarms that were not generated on time within a time period of weeks.

[0087] Calculation data stability:

[0088] SA x =(Cycle x *100),

[0089] Where x is a time period, including but not limited to hours, days, or weeks; the stability of the data is calculated based on whether the data is produced periodically.

[0090] The algorithm for calculating the importance of each data in the data inventory is as follows:

[0091] (1) Calculate the initial self-citation degree of each data in the stock data:

[0092] (αCI n +(1-α)*(DI n +SA x )),

[0093] Where α is the damping coefficient, CI n is the citation degree of the nth data, DI n is the completeness of the nth data, SA x is the stability over x periods, where x is hour, day, or week.

[0094] (2) Add the citations based on the citation status of each data:

[0095]

[0096] Among them, p is the total number of data; Out n is the number of outlinks of the nth data reference node;

[0097] (3) Calculate the importance of data:

[0098]

[0099] Among them, TR is the importance.

[0100] According to the above formula, the TR value of the data can be calculated to obtain the final importance result.

[0101] After calculating the importance of each piece of data in the stock data, the data is ranked based on the importance and a preset number of pieces of data with the highest importance are selected as key data. It should be understood that the preset number can be selected based on different types of data and the present invention is not limited thereto. For example, the top 50% of the total importance of all the stock data can be selected as key data.

[0102] S3. Standardize key data and determine data standards.

[0103] After filtering out the key data, the key data is standardized according to the business theme, data source, data description and other contents in the above data integrity, so as to determine the data standard. The main idea is to normalize the data information, express the data information as a set of all attributes, determine the characteristic attributes of each attribute information in the key data (including but not limited to null values, text descriptions, encoding and other attribute information), form a characteristic attribute set of each attribute information, and then combine the characteristic attribute sets of each attribute information into a standard set, thereby forming a data standard set. Taking the data in the RRP data module as an example, assuming that in the relevant personnel information in the data, the characteristic attributes of each attribute in the key data include name, age and phone number respectively, then the standard set includes the three attributes of name, age and phone number.

[0104] In an optional embodiment of the present invention, for key data to be standardized, classification analysis of the data can be performed through a neural network.

[0105] In an optional embodiment of the present invention, attribute information of key data can be distinguished and attribute information that meets preset redundant information criteria can be deleted. For example, attribute information that is meaningless or has little impact on the work can be deleted to improve the effective expression of attribute information. The purpose of normalizing data is to remove compressed and redundant information in the original data and enhance useful information.

[0106] In an optional embodiment of the present invention, data with the same data type can be clustered based on the data type of the characteristic attributes in the attribute information of the key data (data to be standardized) to generate a corresponding identification system. For example, after combining the characteristic attribute sets of various attribute information, the resulting data standard set is {Device Domain - Source Layer - Table 1 - Temperature - nn°C}, and the corresponding code is xxx-yyyy-zzzz-tem-00nn.

[0107] S4. Generate rules based on data standards.

[0108] Specifically, the rules are generated based on the generated standards, and the rules include naming rules, coding rules and quality characteristic audit rules. Naming rules are used to unify all naming specifications, including but not limited to Chinese naming, English naming, Chinese field naming, and English field naming rules; part of the coding rules is used to constrain the encoding of the value dictionary, and part of it is used to verify the encoding of equipment-related data, such as equipment coding; quality characteristic audit rules are used to verify the integrity, uniqueness, timeliness, validity, accuracy, and consistency of the data, such as non-empty constraints to verify the integrity of data. The rules include the definition of the data format, which includes the definition of the data, the data structure, and the value list. Take the rules related to ERP data as an example to illustrate. For example, after the above-mentioned determination of the data standards, it has been statistically determined that the personnel of the 5 items account for a large proportion. The rules for the personnel information in the ERP data can be the five items including name, age, gender, address, and telephone number. The data format of the three items of name, gender, and address is in text format, while the data format of age and telephone number is in numerical format.

[0109] S5. Verify the existing data and / or standardize the newly added data based on the rules, and use the verified existing data and / or the standardized newly added data as standard data.

[0110] Specifically, for massive amounts of existing data, manually verifying each item individually for compliance with data governance requirements is clearly unrealistic. For example, taking the personnel information data in the ERP data module, while statistics show that 80% of employees have five items in their information, and only 20% have three or more items, identifying these 20% of employees who do not comply with the rules is still a significant workload. Furthermore, the information for the remaining 80% may also not comply with the rules. For example, age may be in text format rather than numeric format. To address this, it is necessary to obtain existing data from each data module, verify it against the rules, and convert it into standardized data that meets the standards. For example, obtain the personnel information data in the ERP data module and verify it item by item. Data that meets the rules is considered standardized data. For example, for the five items of name, age, gender, address, and phone number, where the name, gender, and address are in text format, and the age and phone number are in numeric format, this is considered standardized data.

[0111] Existing data whose formats do not conform to the regulations must be corrected according to the regulations, including but not limited to manual correction and automatic program correction, to obtain standard data. For example, if the phone number and address of a person are missing, they can be manually recorded. If the data format does not meet the requirements, the data format can be automatically converted by a computer program, such as converting text-formatted age data into numeric data.

[0112] New data from different subsystems is often unregulated and has a more chaotic format. This data, when added to existing data, further creates confusion. New data may come from different data modules and have different formats. Non-standardized new data can have disastrous consequences for the enterprise. Therefore, new data must be standardized according to rules to ensure that it complies with standards. This means converting new data into standardized data.

[0113] Continuing with the ERP data module as an example, when there are personnel changes in the company, such as when a new employee joins the company, the new employee information needs to be entered through the ERP system. According to the personnel information rules, for example, the five items including name, age, gender, address, and phone number can be entered. The data format of name, gender, and address is text format, while the data format of age and phone number is numeric format. When entering personnel information, if an item is missing or the numeric format of a certain item does not meet the requirements of the rules, the ERP system can issue an error warning and the erroneous data will not be accepted. Only when the new employee's data meets the various constraints of the rules can this new data be accepted, thereby obtaining standardized new data.

[0114] In an optional embodiment of the present invention, after step S5, the method further includes constructing panoramic data based on the association relationship between the standard data and the data counterparts.

[0115] Specifically, standard data is associated with all relevant data modules to build a panoramic view. The accuracy of the associated data is supported by standards and rules, and the standards and rules are optimized through the usage of panoramic business data association to form a closed loop. Data counterparts include but are not limited to personnel, equipment, products, etc. corresponding to the standard data. For example, for a certain device (i.e., the data counterpart), the standard data corresponding to the device in each data module can be associated, that is, the panoramic data of the device is constructed. The panoramic data includes the data corresponding to the device in each data module, thereby realizing the integration of device data.

[0116] The data governance method provided by the present invention is described in detail below using equipment data from a manufacturing industry as an example:

[0117] like Figure 3 As shown, a manufacturing equipment data management method specifically includes the following steps:

[0118] Step 1: Organize equipment-related business processes and identify equipment-related data from top to bottom. Equipment is a physical entity. For example, a piece of equipment exists within a plant area and a unit within that plant area. The equipment includes a device, which contains a component, which contains a part. The top-down equipment-related data in the equipment business process is represented by the following sequence: plant area - unit - equipment - device - component - part.

[0119] Step 2: Investigate business systems and identify device-related data from the bottom up. For example, business system data is organized into a logical hierarchy, from underlying IoT sensors to data storage, monitoring systems, and system applications. This logical hierarchy represents the device-related data in the business system from the bottom up.

[0120] Step 3: Classify and grade the device-related data based on the identified relevant data and combined with security management and control requirements.

[0121] Step 4: Based on the identified device-related data, standards are established according to the existing data. The standards include but are not limited to naming, data structure, value list, etc. Steps 4, 5, and 6 can be combined and carried out simultaneously.

[0122] Step 5: Based on the content of the standard, formulate rules for generating the standard, verify the already standardized data and constrain the data added in the future, including naming rules, coding rules, quality characteristic rules, etc.

[0123] Step 6: Streamline the structure of the device-related data tables. This step primarily addresses inconsistencies in the same field across different tables. For example, DeviceCode and DeviceNo both represent "device codes" but are identified by different fields. By streamlining the table structure, the fields can be unified. Steps 4, 5, and 6 can be performed simultaneously.

[0124] Step 7: Based on the identified and standardized device-related data, associate the device data to form a panoramic view of the device.

[0125] The manufacturing equipment data governance method described above as an example is a closed-loop approach to manufacturing equipment data governance that utilizes the "establish standards - generate rules - associate data" process. This approach focuses on equipment data governance and can associate data based on different equipment dimensions. For example, the equipment's lifecycle dimension can be divided into design, production, operation, and maintenance, and the equipment data feature dimension can be divided into static data and dynamic data. By establishing standards, existing equipment data is standardized, and rules are generated based on standards to verify future newly added equipment data. Standards and rules support equipment data association, and equipment-associated data can optimize standards and rules. This closed-loop data governance approach standardizes existing data, constrains new data, unifies standard data sources, and improves data quality.

[0126] In summary, the present invention provides a data governance method, which obtains the stock data in at least one data module, filters out the key data in the stock data; standardizes the key data and determines the data standard; generates rules based on the data standard, verifies the stock data and / or standardizes the newly added data in the data module based on the rules, uses the verified stock data and / or the standardized newly added data as standard data, and stores the standardized newly added data in the corresponding data module. Through the implementation of this method and the application of the system, standards are formulated according to the stock data in each data module, and then rules are generated according to the standards. Finally, the stock data in the data module and the newly added data to be stored in the data module are standardized based on the rules, thereby realizing closed-loop governance of data, unifying the interfaces and data formats of each data module, reducing the difficulty of data integration, improving the usability of data, and ensuring the timeliness and accuracy of data maintenance.

[0127] Based on the same inventive concept, the present invention further provides a data governance system, which is used to execute any of the above-mentioned data governance methods. The data governance system provided by the present invention is described below, and the data governance system described below and the data governance method described above can be referenced in conjunction with each other.

[0128] like Figure 4 As shown, the invention provides a data governance system, including a data governance unit 40 and at least one data module, such as a product data management data module 41, an enterprise resource planning data module 42, a production execution system data module 43, an operation and maintenance data module 44, a quality management system data module 45, a warehouse management system data module 46, etc.

[0129] The data governance unit is used to obtain the stock data in at least one data module, screen out key data in the stock data, standardize the key data, determine data standards, generate rules based on the data standards, verify the stock data based on the rules, and standardize the newly added data based on the rules, and use the verified stock data and the standardized newly added data as standard data;

[0130] like Figure 5 As shown, the data governance unit 40 includes a basic data support layer 50, which includes a standard unit 51 and a rule unit 52; the standard unit 51 contains data standards for summarizing relevant data in data modules across various subsystems; the rule unit 52 contains rules for verifying standards, verifying the accuracy of existing data, and standardizing new data.

[0131] Standard unit 51 contains generated standards, including naming standards, data structure standards, and value list standards. Naming standards are used to summarize all cross-system standard naming, including but not limited to table Chinese and English naming, and field Chinese and English naming dictionaries. Data structure standards are used to standardize the structure of device-related data tables, including but not limited to field Chinese and English names, data types, data lengths, value ranges, and entry specifications. Value lists are used to standardize value range entry, ensuring that the value ranges in the data structure are accurate and up-to-date.

[0132] The rule unit 52 contains rules generated from the generated standards, which are used to standardize existing data in each data module and regulate new data. These rules include naming rules, encoding rules, and quality characteristic auditing rules. Standardization includes definition standards, data structure standards, and value list standards. Naming rules are used to standardize all naming conventions, including but not limited to table naming in Chinese and English, and field naming in Chinese and English. Encoding rules are used to constrain the encoding of value dictionaries and verify the encoding of device-related data, such as device codes. Quality characteristic auditing rules are used to verify data integrity, uniqueness, timeliness, validity, accuracy, and consistency, such as verifying data integrity using non-null constraints. The data modules of each subsystem send existing data to the rule unit 52, which verifies the data according to the rules and converts it into standard data that complies with the standards. The rule unit 52 also sends the standard data to the panoramic data construction layer 60. For new data, the rule unit 52 constrains and regulates it. New data may come from different data modules and have different formats. Non-standardized new data can have disastrous consequences for the enterprise. In this regard, the rule unit 52 provides the data module with new data rules to regulate the new data so that the new data also meets the requirements of the standard, that is, the new data is also converted into standard data.

[0133] like Figure 5 and Figure 6 As shown, in an optional embodiment of the present invention, the data governance unit 40 also includes a panoramic data construction layer 60. The panoramic data construction layer 60 associates all data modules based on the standard data obtained by the basic data support layer 50 to form panoramic data. The panoramic data construction layer 60 supports the accuracy of the associated data through standards and rules, and optimizes the standards and rules based on the panoramic data association. The basic data support layer 50 is at the bottom layer, and the panoramic data construction layer 60 is at the upper layer. The bottom basic data support layer 50 provides the upper panoramic data construction layer 60 with governed data, and the panoramic data construction layer 60 constructs panoramic data based on the governed data.

[0134] Specifically, the panoramic data construction layer 60 includes a business combing module 61, a data identification module 62, a data classification and grading module 63, a table structure module 64 and a data association module 65 which are connected in sequence.

[0135] The business combing module 61 is used to implement top-down planning of business data, first combing the first-level business process, and then refining the second-level and third-level business processes on this basis. The data identification module 62 is used to identify equipment-related data to ensure that business entities are not missing. These identified data include PDM data, ERP data, MES data, MRQ data, QMS data, and WMS data. The data classification and grading module 63 is used to realize the rapid management of equipment data as data assets, while dividing the levels to meet security requirements. The table structure module 64 is used to implement the formulation of the table structure of equipment-related data. The data association module 65 is used to clarify the association relationship between the data counterpart and the related data based on the standard data obtained from the basic data support layer, and construct the global data of the data counterpart.

[0136] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0137] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A data governance method, characterized in that: include: Obtaining stock data in at least one data module; Filtering out key data from the stock data; Standardize the key data and determine data standards; generating rules based on the data standards; Verifying the existing data and / or standardizing the newly added data based on the rules, and using the verified existing data and / or the standardized newly added data as standard data; Filter out key data from the stock data, including: Calculating the importance of each data in the stock data; sorting the data based on the importance; Select the preset number of data in the front as key data; Calculating the importance of each data in the stock data includes: Calculate the initial citation of each data in the stock data: ； in, α is the damping coefficient, For the n The citation of data, For the n The completeness of the data, for x Stability under cycles, x is the time period; Superimpose citations based on the data's citation status: ； in, is the total number of data; is the number of outlinks of the nth data reference node; Calculate the importance of data: ； in, TR For importance; Iterative calculation is performed until the importance tends to be stable, and the final importance result is obtained.

2. The data governance method according to claim 1, characterized in that: The calculation of the citation degree includes: ； in, For the n Number of queries for a service; For the n The number of downstream users of a service; The stability calculation includes: Calculation statistical period: ； in, is the statistical period in hours. is the statistical period with days as the time unit, is the statistical period with weeks as the time unit, The number of alarms that were not generated on time within a time period in hours. The number of alarms that were not generated on time within a time period in days. The number of alarms that were not generated on time within a time period of weeks; Computational stability: 。 3. The data governance method according to claim 1, wherein: Standardize the key data and determine the data standards, including: Distinguishing the attribute information of the key data and deleting the attribute information that meets the preset redundant information conditions; Determine characteristic attributes of each attribute information in the key data, and construct a characteristic attribute set of each attribute information; The characteristic attribute sets of the various attribute information are combined to construct a data standard set.

4. The data governance method according to claim 3, characterized in that: Generating rules based on the data standards includes: Based on the data standard set, a rule including a data format is generated; the data format includes a data definition, a data structure, and a value list.

5. The data governance method according to claim 4, characterized in that: Verifying the existing data and / or standardizing the newly added data based on the rules includes: Verifying the stock data in the data module one by one according to the rules, screening out the stock data whose data formats comply with the rules as standard data, correcting the screened stock data whose data formats do not comply with the rules according to the rules, and using the corrected stock data as standard data; And / or, the newly added data is verified according to the rule, and the newly added data whose data format complies with the rule is screened out as standard data.

6. The data governance method according to any one of claims 1 to 5, characterized in that: After the verified stock data and / or the standardized newly added data are used as standard data, the method further includes: Panoramic data is constructed based on the association relationship between the standard data and the data counterparts.

7. A data governance system, characterized in that: for executing the data governance method according to any one of claims 1 to 6, the system comprising a data governance unit and at least one data module, the data module containing stock data; the data governance unit being configured to obtain the stock data in the at least one data module, screen out key data in the stock data, perform standardization on the key data, determine data standards, generate rules based on the data standards, verify the stock data based on the rules, and standardize newly added data based on the rules, and use the verified stock data and the standardized newly added data as standard data; The data governance unit includes a basic data support layer, and the basic data support layer includes a standard unit and a rule unit; the standard unit contains the data standard; and the rule unit contains the rule.

8. The data governance system according to claim 7, wherein: The data governance unit also includes a panoramic data construction layer, which is used to construct panoramic data based on the association relationship between the standard data and the data counterparts.

Images