A method for constructing an enterprise data platform

Abstract

The present invention provides a method for constructing an enterprise data platform, including: establishing a unified enterprise model UEM; matching the unified enterprise model UEM with the actual business activities to obtain a target enterprise model EM of the target enterprise; obtaining an operation process data node BPN; obtaining the data analysis requirements of each operation process data node BPN; and establishing a detailed business data table DBDT unique to the operation subject for each operation process data node BPN. The EDP constructed using the pre-constructed standardized and reusable DBDT can quickly meet various possible business analysis requirements and keep the data caliber in various analysis results unique, thereby comprehensively improving the data service capabilities of the EDP. Since the DBDT can be directly used to form DM layer analysis indicators, the development process is simplified and the development efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of enterprise data processing, and in particular relates to a method for constructing an enterprise data platform. Background Art

[0002] The advent of the data age has resulted in an explosive growth of data, which has brought unprecedented challenges and opportunities to enterprises. Building processing and analysis capabilities for large amounts of data and realizing the value of data are gradually becoming one of the core elements for improving the competitiveness of enterprises. Therefore, the Enterprise Data Process Platform (EDP) has quickly become an important means for enterprises to process and analyze data. Depending on its structure and function, the enterprise data platform can be manifested as a data warehouse, data lake, lake-warehouse integrated platform, data middle platform, etc. The core function of the enterprise data platform is to collect, aggregate and process the data generated by the enterprise, and then provide convenient, fast and high-quality data analysis and services.

[0003] In the prior art, enterprise data platforms have the following problems:

[0004] (1) Due to the complexity of enterprise business activities, there is no unified method for business division of data and method for predicting analysis needs. Therefore, faced with a large amount of data generated by enterprise business activities, there is a lack of effective methods for pre-integration and processing of data from a business perspective. As a result, the analysis needs can only be responded to passively, making it difficult to meet the data analysis needs from the business in a predictable and prepared manner.

[0005] (2) The enterprise data platform needs to establish a specific detailed business data table for each data analysis requirement. After the development for a specific data analysis requirement is completed, when a new data analysis requirement arises, the previous detailed business data table cannot be reused and a new detailed business data table needs to be redeveloped from the bottom up, resulting in a cumbersome development process and low development efficiency. Summary of the invention

[0006] In view of the defects in the prior art, the present invention provides a method for constructing an enterprise data platform, which can effectively solve the above problems.

[0007] The technical solution adopted by the present invention is as follows:

[0008] The present invention provides a method for constructing an enterprise data platform, comprising the following steps:

[0009] Step S1, establishing a unified enterprise model UEM; the unified enterprise model UEM is used to describe the rules for decomposing enterprise business activities, thereby decomposing enterprise business activities into interrelated elements at different levels from different perspectives;

[0010] Step S2, collecting all actual business activities of the target enterprise; matching the unified enterprise model UEM with the actual business activities to obtain the target enterprise model EM of the target enterprise, the target enterprise model EM includes an operation activity decomposition structure and a management activity decomposition structure; the operation activity decomposition structure describes the business domain-operation process-operation activity and their hierarchical relationship formed after all actual business activities of the target enterprise are decomposed; the management activity decomposition structure describes the business domain-management subject-management target-management element and their hierarchical relationship formed after all actual business activities of the target enterprise are decomposed;

[0011] Step S3, comprehensively analyzing the target enterprise's operation activity decomposition structure and management activity decomposition structure, and obtaining the operation activity required for data analysis of each management element in the business domain, and the operation activity is called the operation process data node BPN of the management element;

[0012] Construct the data analysis requirements of each corresponding operation process data node BPN of the management element, and obtain the BPN required by the management element for data analysis, the required data analysis requirement table DN, the data encryption policy table DEP and the privacy data processing policy table PDP;

[0013] Step S4, summarizing the data analysis requirements of all business domains of the target enterprise according to the operation process data node BPN, obtaining all data analysis requirements for each operation process data node BPN, and forming a data analysis requirement table DN, a data encryption policy table DEP, and a privacy data processing policy table PDP belonging to the operation process data node BPN;

[0014] Step S5: For each operation process data node BPN, a detailed business data table DBDT unique to the operation subject is established:

[0015] For each job process data node BPN, multiple job topics of the next level contained in the job process data node BPN are generated according to the management attributes; for each job topic, its data analysis requirements are associated, and the required data analysis requirement table DN, data encryption policy table DEP and privacy data processing policy table PDP are constructed;

[0016] Construct a detailed business data table DBDT uniquely corresponding to each job subject, so that the detailed business data table DBDT set formed by all job subjects under the same job process data node BPN at least includes all data elements in the data analysis requirement table DN under the job process data node BPN, and all policies in the data encryption policy table DEP and the privacy data processing policy table PDP under the job process data node BPN;

[0017] Step S6, establishing a dynamic DBDT management directory; storing the DBDT management directory and all detailed business data tables DBDT of the target enterprise into the DWR data report layer;

[0018] At this point, before developing for data analysis needs, the enterprise data platform EDP including the target enterprise model EM of the target enterprise and the DWR data report layer of all detailed business data tables DBDT is pre-built;

[0019] The enterprise data platform EDP is used to develop data analysis services for various data analysis needs from various business domains.

[0020] Preferably, step S1 specifically comprises:

[0021] The unified enterprise model UEM includes a business domain model, an operation process reference model, a management element reference model and a business domain-management subject-operation process comparison table;

[0022] The business domain model is used to define the rules for dividing the business activities of an enterprise into multiple business domains, thereby dividing the business activities of an enterprise into a number of business domains that are independent but interconnected, classified, and have a mutual support and driving relationship, and describes the superior-subordinate relationship between each of the business domains. The divided business domains are applicable to various types of target enterprises;

[0023] The operation process reference model is used to define the rules for decomposing the business activities of an enterprise from the operation perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-operation process-operation activity". Therefore, for the business activities of each business domain, the operation process reference model is used to decompose the corresponding operation process, and then the operation process is decomposed into multiple operation activities. The decomposed operation processes and operation activities include the operation processes and operation activities of various types of target enterprises.

[0024] The management element reference model is used to define the rules for decomposing the business activities of an enterprise from a management perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-management theme-management goal-management element". Therefore, for the business activities of each business domain, the management goal under the set management theme is further decomposed into multiple management elements, that is, the management elements that have a direct impact and driving relationship with the corresponding management theme and management goal. The decomposed management themes, management goals and management elements include the management themes, management goals and management elements of various types of target enterprises.

[0025] The business domain-management subject-operation process comparison table is used to describe the association relationship between the operation process reference model and the management element reference model. Specifically, in the business domain, the management subject and the operation process are matched, and then the management element of the next level of the management target of the next level of the management subject and the operation activity of the next level of the operation process are established.

[0026] Preferably, step S2 specifically comprises:

[0027] Step S2.1, inputting each actual business activity of the target enterprise into the unified enterprise model UEM;

[0028] Step S2.2, using the business domain model of the unified enterprise model UEM, analyzing the actual business activities, and dividing the actual business activities into business domains of multiple target enterprises based on the rules of UEM;

[0029] Step S2.3, input each of the business domains and the actual business activities into the operation process reference model; the operation process reference model decomposes the actual business activities and the business activities planned to be carried out in the future from the operation perspective, and obtains the target enterprise operation process and operation activities with a hierarchical structure of business domain-operation process-operation activity, which is the operation activity decomposition framework;

[0030] Step S2.4, input each of the business domains and the actual business activities into the management element reference model; the management element reference model decomposes the actual business activities from a management perspective to obtain a target enterprise management theme, management goal and management element with a hierarchical structure of business domain-management theme-management goal-management element, which is the management activity decomposition framework.

[0031] Preferably, step S3 specifically comprises:

[0032] Step S3.1, based on the business domain-management theme-operation process comparison table, comprehensively analyze the operation activity decomposition structure and management activity decomposition structure of the target enterprise, and obtain the operation activity that needs to provide data for data analysis of each management element in the business domain, that is, the operation process data node BPN of the management element;

[0033] Step S3.2, construct the data analysis requirement table DN:

[0034] Each operation activity generates multiple data objects; after a data object is structured, it is a data table, and the data table has multiple data elements, and one data element is a field; the data table is stored in the enterprise data platform;

[0035] For each management element, the constructed data analysis requirement table DN includes the following fields: business domain, management subject, management goal, management element, required BPN, required measurement data, and required dimension data; the required measurement data refers to the measurement data extracted from the data elements in the required data object; the required dimension data refers to the dimensions formed by some or all of the data elements in the required data object;

[0036] Step S3.3, construct the data encryption policy table DEP:

[0037] For each management element, a data encryption policy table DEP is constructed, including the following fields: business domain, management subject, management target, management element, required BPN, required data object, required data element, condition and encryption policy; its meaning is: when the value of a data object or a data element of a certain BPN required by a certain management element meets a certain condition, the corresponding encryption policy needs to be adopted for the data;

[0038] Step S3.4, construct the privacy data processing policy table PDP:

[0039] For each management element, a privacy data processing policy table PDP is constructed, including the following fields: business domain, management subject, management objective, management element, required BPN, required data object, required data element, condition and privacy data processing strategy; its meaning is: when the value of a data object or a data element of a BPN required by a management element meets a certain condition, the corresponding privacy data processing strategy needs to be adopted for the data.

[0040] Preferably, in step S5, the detailed business data table DBDT uniquely corresponding to the operation subject includes a DBDT master table, a data encryption policy table DEP and a privacy data processing policy table PDP;

[0041] The DBDT master table includes the following fields: business domain, operation process, operation activity, operation theme, measurement, modification and dimension, which means: for a certain operation theme under the business domain, operation process and operation activity, the corresponding operation data is stored in the form of measurement, modification and dimension;

[0042] The data encryption policy table DEP and the privacy data processing policy table PDP are used to define the data encryption policy and privacy data processing policy to be adopted when the DBDT main table provides corresponding job data.

[0043] Preferably, the detailed business data table DBDT uniquely corresponding to the operation theme established in step S5 becomes a standardized and reusable DBDT because it contains data required for all possible analysis requirements of various management elements in all business domains of the target enterprise.

[0044] Preferably, it also includes:

[0045] Step S7, manage and maintain the uniqueness of each detailed business data table DBDT based on the DBDT management directory; when it is necessary to adjust and modify the established detailed business data table DBDT, trace back and find the relevant indicators and data reports that have referenced the detailed business data table DBDT according to the data reference relationship, and modify the existing indicators and data reports according to the data logic of the modified detailed business data table DBDT.

[0046] The present invention also provides an enterprise data platform constructed by the construction method of the enterprise data platform, comprising: a target enterprise model EM, an ODS source data layer, a DWI data integration layer, a DWR data report layer and a DM data mart layer;

[0047] The target enterprise model EM is used to describe the rules and results of decomposition of the target enterprise's business activities, thereby decomposing the enterprise's business activities into interrelated elements at different levels from different perspectives;

[0048] The ODS source data layer is used to collect and store various types of original business activity data generated by enterprise business activities in an event-driven manner;

[0049] The DWI data integration layer is used to clean and integrate the original business activity data of the ODS source data layer, and then obtain a list of operating activities matching the target enterprise model EM according to the target enterprise model EM, form an operating process data node BPN, and store it;

[0050] The DWR data report layer is used to establish and store a unique detailed business data table DBDT for multiple operation themes according to data analysis requirements;

[0051] The DM data mart layer is used to construct a DM layer data model based on the detailed business data table DBDT unique to multiple operation themes of the DWR data report layer; based on the DM layer data model, design and develop data mart DM data generated through indicator and dimension design to provide a data basis for various data analyses of the enterprise.

[0052] The method for constructing an enterprise data platform provided by the present invention has the following advantages:

[0053] The EDP constructed by the present invention using the pre-built standardized and reusable DBDT can quickly meet various possible business analysis needs and keep the data caliber in various analysis results unique, thereby comprehensively improving the data service capabilities of the EDP. Since the DBDT can be directly used to form DM layer analysis indicators, the development process is simplified and the development efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of a process for constructing an enterprise data platform provided by the present invention;

[0055] Figure 2 An example diagram of a process for summarizing data analysis requirements from various business domains provided by the present invention. DETAILED DESCRIPTION

[0056] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0057] The present invention provides a method for constructing an enterprise data platform (EDP) to achieve the following three objectives:

[0058] 1) Pre-build a unified enterprise model UEM that includes complete management concerns and complete business activities in different business areas. The UEM is suitable for different types of target enterprises, and the unified enterprise model UEM is matched with actual enterprise business activities to pre-derive BPN for each operation process data node and possible future data analysis needs in the absence of data analysis needs from actual business departments.

[0059] 2) For each operation process data node BPN, summarize the possible data analysis requirements from different business fields, and design and develop a unique and "standardized" detailed business data table DBDT for the DWR layer operation theme in the enterprise data platform according to the pre-set rules, so that the detailed business data table DBDT can meet all possible data analysis requirements from different business fields in the future, thereby achieving the purpose of "reusability" of DBDT.

[0060] 3) When actual business data analysis needs arise, design and develop corresponding DM layer data and application analysis reports based on the "standardized and reusable" DBDT pre-built in the DWR layer. Since the "standardized and reusable" DWR layer DBDT can be used directly to form DM layer analysis indicators, it avoids the cumbersome process of developing dedicated detailed business data tables DBDT layer by layer from the DWI layer and even ODS upwards for specific analysis needs every time, which simplifies the development process and improves development efficiency; at the same time, using standardized DBDT each time largely avoids the problems caused by different data calibers.

[0061] Therefore, the present invention pre-processes a standardized and reusable DWR layer DBDT before there is an actual data analysis demand, which can fundamentally play the role of EDP in fast data processing and coordinated caliber to effectively provide data analysis services for enterprises.

[0062] See also Figure 1 The present invention provides a method for constructing an enterprise data platform, in which a standardized and reusable detailed business data table DBDT is pre-established in the DWR data report layer of the enterprise data platform EDP, so that the constructed enterprise data platform EDP can provide efficient data analysis services for the enterprise, including the following steps:

[0063] Step S1, establishing a unified enterprise model UEM;

[0064] Step S2, matching the unified enterprise model UEM with the actual business activities to obtain the target enterprise model EM of the target enterprise;

[0065] Step S3, constructing the data requirements of the management elements for each corresponding operation process data node BPN;

[0066] Step S4, summarizing the data analysis requirements of all business domains of the target enterprise according to the operation process data node BPN;

[0067] Step S5, creating a detailed business data table DBDT unique to each job subject for each job process data node BPN;

[0068] Step S6, establishing a dynamic DBDT management directory;

[0069] Step S7: Perform uniqueness management and maintenance on each detailed business data table DBDT based on the DBDT management directory.

[0070] The following describes each step in detail:

[0071] Step S1, establishing a unified enterprise model UEM;

[0072] The unified enterprise model (UEM) is used to describe the rules for decomposing enterprise business activities, thereby decomposing enterprise business activities into interrelated elements at different levels from different perspectives;

[0073] Specifically, EDP needs to classify, identify, clean, integrate, load, and aggregate the large amount of enterprise business activity data collected from different sources, and process them into data components of different forms and serving different purposes, so as to provide fast data analysis services to external data demanders in a convenient and standardized manner.

[0074] Therefore, the present invention pre-establishes a universal, standardized a priori model that can cover all business activities (including operational activities and management activities) of various types of target enterprises, and classifies, identifies and aggregates the collected business activity data through the a priori model. This model that includes business activities in various fields and integrates the contents of different business activities in various fields is called a unified enterprise model (UEM).

[0075] The unified enterprise model UEM includes a business domain model, an operation process reference model, a management element reference model and a business domain-management subject-operation process comparison table;

[0076] The business domain model is used to define the rules for dividing the business activities of an enterprise into multiple business domains, thereby dividing the business activities of an enterprise into a number of business domains that are independent but interconnected, classified, and have a mutual support and driving relationship, and describes the superior-subordinate relationship between each of the business domains. The divided business domains are applicable to various types of target enterprises;

[0077] The operation process reference model is used to define the rules for decomposing the business activities of an enterprise from the operation perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-operation process-operation activity"; therefore, for the business activities of each business domain, the operation process reference model is used to decompose the corresponding operation process, and then the operation process is decomposed into multiple operation activities. The decomposed operation processes and operation activities include the operation processes and operation activities of various types of target enterprises. Specifically, by summarizing the operation processes of various business domains of various types of enterprises according to the "business domain-operation process-operation activity" model, and then summarizing the operation activities of each operation process, the model and content of business domain-operation process-operation activity can be constructed, and the operation activities carried out according to this model and content can achieve the goal of optimal business performance.

[0078] The management element reference model is used to define the rules for decomposing the business activities of an enterprise from a management perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-management theme-management goal-management element"; therefore, for the business activities of each business domain, for the management goals under the set management theme, multiple management elements can be further decomposed to obtain management elements that have a direct impact and driving relationship with the corresponding management theme and management goal. The decomposed management themes, management goals and management elements include the management themes, management goals and management elements of various types of target enterprises. Specifically, by summarizing the management content of each business domain according to the "business domain-management theme-management goal-management element" model, the management elements in each business domain that have a direct impact and driving relationship with the management goals under each management theme are constructed, that is, all management elements can be monitored according to this model, and management activities can be carried out according to the monitoring results, and the purpose of management goals can be finally achieved.

[0079] The business domain-management subject-operation process comparison table is used to describe the association relationship between the operation process reference model and the management element reference model. Specifically, in the business domain, the management subject and the operation process are matched, and then the management element of the next level of the management target of the next level of the management subject and the operation activity of the next level of the operation process are established.

[0080] Step S2, matching the unified enterprise model UEM with actual business activities:

[0081] Collect all actual business activities of the target enterprise; match the unified enterprise model UEM with the actual business activities to obtain the target enterprise model EM of the target enterprise, the target enterprise model EM includes an operation activity decomposition structure and a management activity decomposition structure; the operation activity decomposition structure describes the content and hierarchical relationship of the business domain-operation process-operation activity formed after all actual business activities of the target enterprise are decomposed; the management activity decomposition structure describes the content and hierarchical relationship of the business domain-management subject-management target-management element formed after all actual business activities of the target enterprise are decomposed;

[0082] Step S2 is specifically as follows:

[0083] Step S2.1, inputting each actual business activity of the target enterprise into the unified enterprise model UEM;

[0084] Step S2.2, using the business domain model of the unified enterprise model UEM, analyzing the actual business activities, and dividing the actual business activities into business domains of multiple target enterprises based on the rules of UEM;

[0085] Step S2.3, input each of the business domains and the actual business activities into the operation process reference model; the operation process reference model decomposes the actual business activities and the business activities planned to be carried out in the future from the operation perspective, and obtains the target enterprise's operation process and operation activities with a hierarchical structure of business domain-operation process-operation activity, which is the operation activity decomposition framework of the target enterprise;

[0086] Step S2.4, input each of the business domains and the actual business activities into the management element reference model; the management element reference model decomposes the actual business activities from a management perspective to obtain a target enterprise management theme, management goal and management element with a hierarchical structure of business domain-management theme-management goal-management element, which is the management activity decomposition framework of the target enterprise.

[0087] When performing this step, another possible method is:

[0088] Each actual business activity is pre-decomposed into multiple elements such as business domain, operation process, and operation activity according to the rules defined by UEM.

[0089] Then, the decomposed elements are benchmarked according to business domains and the unified enterprise model (UEM):

[0090] ① Identify the elements consistent with UEM according to the names and numbers of the corresponding elements of UEM;

[0091] ② Add the elements that are not in UEM to the corresponding positions under the corresponding business domain of UEM, and assign new names and numbers according to the rules of UEM;

[0092] ③ Elements that are not matched in UEM but that the enterprise still hopes to develop in the future;

[0093] ④ Delete the remaining elements in UEM.

[0094] ⑤Summarize the above results and obtain the "target enterprise model EM", including business domain, operation process, and operation activities.

[0095] The content of this target enterprise model EM is the work activity decomposition framework obtained from the perspective of UEM.

[0096] ⑥ Summarize the business domains, operating processes, and operating activities contained in the target enterprise model EM, and match the business domains, management themes, management objectives, and management elements contained in the target enterprise model EM according to the business domain-management theme-operating process comparison table and the management element reference model to obtain the target enterprise management activity decomposition framework.

[0097] Step S3, construct management elements for data analysis requirements of each corresponding operation process data node BPN:

[0098] Comprehensively analyze the target enterprise's activity decomposition structure and management activity decomposition structure to obtain the activity that needs to provide data for each management element in the business domain when performing data analysis. This activity is called the operation process data node BPN of the management element.

[0099] Construct the data analysis requirements of each corresponding operation process data node BPN of the management element, and obtain the BPN required by the management element for data analysis, the required data analysis requirement table DN, the data encryption policy table DEP and the privacy data processing policy table PDP;

[0100] Step S3 is specifically as follows:

[0101] Step S3.1, based on the business domain-management theme-operation process comparison table, comprehensively analyze the operation activity decomposition structure and management activity decomposition structure of the target enterprise, and obtain the operation activity that needs to provide data for data analysis of each management element in the business domain, that is, the operation process data node BPN of the management element;

[0102] Specifically, taking the business domain as the unit, we can obtain the work activities that need to provide data when conducting business analysis on each management element. By summarizing such work activities in all business domains of the target enterprise, we can obtain a list of all work activities that may provide data required for business analysis. Each work activity in the work activity list is called a business process data node (BPN).

[0103] Step S3.2, construct the data analysis requirement table DN:

[0104] Each operation activity generates multiple data objects; after a data object is structured, it is a data table, and the data table has multiple data elements, and one data element is a field; the data table is stored in the enterprise data platform;

[0105] For each management element, the data analysis requirement table DN is constructed, including the following fields: business domain, management subject, management objective, management element, required BPN, required measurement data and required dimensional data; among which, the required measurement data refers to the measurement data extracted from the data elements in the required data object; the required dimensional data refers to the dimensions formed by some or all of the data elements in the required data object.

[0106] The table structure of the data analysis requirement table DN is as follows:

[0107]

[0108] Step S3.3, construct the data encryption policy table DEP:

[0109] For each management element, a data encryption policy table DEP is constructed, including the following fields: business domain, management subject, management objective, management element, required BPN, required data object, required data element, condition and encryption policy; its meaning is: when the value of a data object or a data element of a certain BPN required by a certain management element meets a certain specific condition, the corresponding encryption strategy needs to be adopted for the data.

[0110] The table structure of the data encryption policy table DEP is as follows:

[0111]

[0112] Step S3.4, construct the privacy data processing policy table PDP:

[0113] For each management element, a privacy data processing policy table PDP is constructed, including the following fields: business domain, management subject, management objective, management element, required BPN, required data object, required data element, condition and privacy data processing strategy; its meaning is: when the value of a data object or a data element of a BPN required by a management element meets a certain condition, the corresponding privacy data processing strategy needs to be adopted for the data.

[0114] The table structure of the privacy data processing policy table PDP is as follows:

[0115]

[0116] The implementation process of step S3.3 to step S3.4 can be as follows: taking the business domain as a unit, determine the data security policy (including data encryption policy and privacy data processing policy) required for each management element in the business domain to perform data analysis, and obtain the encryption policy set of the data required for analysis of each management element in the business domain, recorded as DEP i (DataEncryption Policy for management factor i) and the privacy data processing policy set for the data required for each management factor analysis in the business domain, denoted as PDP i (Privacy Data processing Policy for management factor i).

[0117] Step S4, summarizing the data analysis requirements of all business domains of the target enterprise according to the operation process data node BPN, obtaining all data analysis requirements for each operation process data node BPN, and forming a data analysis requirement table DN, a data encryption policy table DEP, and a privacy data processing policy table PDP belonging to the operation process data node BPN;

[0118] Specifically, traverse all business domains of the target enterprise and obtain the DN of each business domain i 、DEP i 、PDP i , summarize the DN of all business domains obtained after traversal i 、DEP i 、PDP i , and then classified according to the BPN caliber, we can obtain the data analysis requirements and security policies from the perspective of BPN for each BPN and from all business domains of the target enterprise. Based on this, we can obtain the complete data requirements for each BPN.

[0119] Taking the production process completion BPN as an example, the process of summarizing the data analysis requirements from various business domains is as follows: Figure 2 shown.

[0120] The specific solution process for the complete data analysis requirements of all management elements in each business domain is as follows:

[0121] Let N be the maximum value of the business domain sequence number in the target enterprise model EM, and N is a natural number.

[0122] Set variable BF j (Business Field, i.e., business domain, j is a natural number from 1 to N), when j = 1, BF j represents business domain 1; when j = 2, BF j represents service domain 2; ..., when j = N, BF j Represents business domain N.

[0123] Set the set DN, which means: the data requirement result set of the management element for the job process data node BPN.

[0124] Set the set DEP, which means: manage the data encryption policy result set of the element for the job process data node BPN.

[0125] Set the set PDP, which means: the privacy data processing policy result set of the management element for the job process data node BPN.

[0126] Set the collection DN j , which means: the data requirements of the relevant operation process data nodes BPN involved in all management elements in business domain j.

[0127] Set Collection DEP j , which means: the data encryption strategy of the relevant operation process data nodes BPN involved in all management elements in business domain j.

[0128] Set up collective PDP j , which means: the privacy data processing strategy of the relevant operation process data nodes BPN involved in all management elements in business domain j.

[0129] Let j = 0;

[0130] DN j =DEP j =PDP j =Φ;

[0131] Let DN = DN j , DEP=DEP j ,PDP=PDP j ;

[0132] Solving process:

[0133] Let j = j + 1;

[0134] Apply the method of step S3.2 and step S3.3, based on BF j , get DN j 、DEP j and PDP j ;

[0135] DN=DN∪DN j , DEP=DEP∪DEP j , PDP=PDP∪PDP j ;

[0136] Determine whether j is greater than N. If not, continue the solution process; if so, end the solution process.

[0137] Finally, the obtained DN, DEP, and PDP are the complete data analysis requirements for all management elements in each business domain.

[0138] Step S5: For each operation process data node BPN, a detailed business data table DBDT unique to the operation subject is established:

[0139] For each job process data node BPN, multiple job topics of the next level contained in the job process data node BPN are generated according to the management attributes; for each job topic, its data analysis requirements are associated, and the required data analysis requirement table DN, data encryption policy table DEP and privacy data processing policy table PDP are constructed.

[0140] Construct a detailed business data table DBDT uniquely corresponding to each job subject, so that the detailed business data table DBDT set formed by all job subjects under the same job process data node BPN at least includes all data elements in the data analysis requirement table DN under the job process data node BPN, and all policies in the data encryption policy table DEP and the privacy data processing policy table PDP under the job process data node BPN;

[0141] Based on the above steps, a detailed business data table DBDT uniquely corresponding to the operation subject can be obtained.

[0142] The detailed business data table DBDT uniquely corresponding to the job subject consists of three parts: the DBDT main table, the data encryption policy table DEP and the privacy data processing policy table PDP; the table structure of the data encryption policy table DEP and the privacy data processing policy table PDP is the same as the table structure of the aforementioned steps S3.3 and S3.4.

[0143] The DBDT main table includes the following fields: business domain, operation process, operation activity, operation theme, measurement, modification and dimension, which means: for a certain operation theme under the business domain, operation process and operation activity, the corresponding operation data is stored in the form of measurement, modification and dimension; wherein, measurement, modification and dimension can contain multiple fields. Measurement is the core indicator, modification is the local dimension, and dimension is the global dimension.

[0144] The table structure of the DBDT main table is as follows:

[0145] Business Domain Operation process Homework activities Assignment Topic measure Modification Dimensions

[0146] The data encryption policy table DEP and the privacy data processing policy table PDP are used to define the data encryption policy and privacy data processing policy to be adopted when the corresponding job data is provided by the DBDT main table.

[0147] In this step, the detailed business data table DBDT uniquely corresponding to the job theme is established, which is a pre-built, standardized and reusable DBDT. Therefore, before developing for data analysis needs, the detailed business data table DBDT is developed and established in the DWR layer of EDP in advance. When the detailed business data table DBDT is used to support DM and the data analysis process thereon, the DBDT master table provides data, and the related data encryption policy table and privacy data processing policy table are responsible for providing the security policy used by the analysis service.

[0148] At this point, a standardized detailed business data table DBDT is obtained. Since the detailed business data table DBDT is designed with all possible business analysis data requirements from various business domains in mind, it can be repeatedly "reused" for various future data analysis requirements.

[0149] For example, in the production completion BPN, in order to meet the two different needs of production completion analysis and resource consumption analysis, DBDTs with two different themes, process completion output and process material consumption, are established respectively.

[0150] The structure of the DBDT master table of the completed process output is as follows:

[0151]

[0152] Step S6, establishing a dynamic DBDT management directory; storing the DBDT management directory and all detailed business data tables DBDT of the target enterprise into the DWR data report layer;

[0153] At this point, before developing for data analysis needs, the enterprise data platform EDP including the target enterprise model EM of the target enterprise and the DWR data report layer of all detailed business data tables DBDT is pre-built;

[0154] The enterprise data platform EDP is used to develop data analysis services for various data analysis needs from various business domains.

[0155] A dynamic DWR layer DBDT management directory is established in EDP to facilitate EDP managers to timely grasp the actual situation of DBDT through query.

[0156] At the same time, the reference relationship between all indicators and data reports developed based on DBDT and the data in DBDT (including the data lineage path of related data in the DM layer) is dynamically recorded for query management.

[0157] Step S7, performing unique management and maintenance on each detailed business data table DBDT based on the DBDT management directory;

[0158] In the daily management of EDP, managers manage and maintain the uniqueness of each detailed business data table DBDT based on the DBDT management directory.

[0159] When it is necessary to adjust and modify the established detailed business data table DBDT, trace back and find the related indicators and data reports that have referenced the detailed business data table DBDT according to the data reference relationship, and modify the existing indicators and data reports according to the data logic of the modified detailed business data table DBDT to ensure the normal availability of the indicators and data reports in use.

[0160] The present invention also provides an enterprise data platform constructed by adopting the construction method of the enterprise data platform, comprising: a target enterprise model EM of the target enterprise, an ODS source data layer, a DWI data integration layer, a DWR data report layer and a DM data mart layer;

[0161] The target enterprise model EM is used to describe the rules and results of decomposition of the target enterprise's business activities, thereby decomposing the enterprise's business activities into interrelated elements at different levels from different perspectives;

[0162] The ODS source data layer is used to collect and store various types of original business activity data generated by enterprise business activities in an event-driven manner;

[0163] Specifically, various types of external data (including IT system data, IOT system data, Internet data, exchange data, manual data, etc.) can be directly collected into the ODS layer of EDP through batch collection, streaming collection, event-driven collection, etc.

[0164] The DWI data integration layer is used to clean and integrate the original business activity data of the ODS source data layer, and then obtain a list of operating activities matching the target enterprise model EM according to the target enterprise model EM, form an operating process data node BPN, and store it;

[0165] Specifically, complete, clean, and single actual business data is obtained through deduplication, transcoding, merging, error correction, aggregation, interpolation, and elimination, and these data are aggregated under each work activity in the work activity list that matches the target enterprise model EM to form integrated DWI layer business data.

[0166] The DWR data report layer is used to establish and store a unique detailed business data table DBDT for multiple operation themes according to data analysis requirements;

[0167] The DM data mart layer is used to construct a DM layer data model based on the detailed business data table DBDT unique to multiple operation themes of the DWR data report layer; based on the DM layer data model, design and develop data mart DM data generated through indicator and dimension design to provide a data basis for various data analyses of the enterprise.

[0168] The enterprise data platform EDP constructed by the present invention has a system function summary as follows:

[0169] 1) UEM management module: used to manage the unified enterprise model UEM and the enterprise model that is benchmarked against the actual enterprise business activities, that is, the target enterprise model EM of the target enterprise.

[0170] 2) Analysis requirements management module: used to manage the analysis requirements of the target enterprise, and decompose and summarize the analysis requirements into the corresponding BPN using the target enterprise model EM.

[0171] 3) DBDT management module: Decompose the data requirements summarized in BPN to form a DBDT that uniquely corresponds to the job subject, uniformly manage and maintain the DBDT, and perform data lineage analysis.

[0172] 4) ETL module: used to collect external data and clean and process data between various data layers within EDP.

[0173] 5) Data model management: Combined with the ETL module, it manages and maintains data models such as indicators.

[0174] 6) System integration module: used to implement and manage the integration and data exchange between EDP and external systems.

[0175] The present invention provides a method for constructing an enterprise data platform, which has the following characteristics:

[0176] 1) By using the unified enterprise model (UEM) and benchmarking it against actual enterprise business activities, the data requirements for all possible business analyses of the enterprise and all BPNs that may need to provide data can be derived.

[0177] 2) A standardized and reusable DBDT can be designed.

[0178] 3) EDP built using standardized and reusable DBDT can quickly meet various possible business analysis needs.

[0179] 4) Rapidly implement data analysis while maintaining the uniqueness of data caliber, thereby giving full play to the role of EDP in promoting the value of enterprise data.

[0180] The method for constructing an enterprise data platform provided by the present invention has the following advantages:

[0181] 1) Make the detailed data required for business analysis completely focus on possible analysis needs, build the corresponding detailed data in advance, and improve the development efficiency of data analysis.

[0182] 2) Maintaining consistency in the scope of the data analysis process can fully meet various business analysis needs while comprehensively improving EDP's data service capabilities and strongly supporting the company's digital transformation.

[0183] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be considered as the scope of protection of the present invention.

Claims

1. A method for constructing an enterprise data platform, characterized in that: The following steps are involved: Step S1, establishing a unified enterprise model UEM; The unified enterprise model (UEM) is used to describe the rules for decomposing enterprise business activities, thereby decomposing enterprise business activities into interrelated elements at different levels from different perspectives; Step S2, collecting all actual business activities of the target enterprise; Matching the unified enterprise model UEM with the actual business activities to obtain a target enterprise model EM of the target enterprise, wherein the target enterprise model EM includes an operation activity decomposition structure and a management activity decomposition structure; The operation activity decomposition framework describes the business domain-operation process-operation activity and their hierarchical relationship formed after all actual business activities of the target enterprise are decomposed; the management activity decomposition framework describes the business domain-management subject-management goal-management element and their hierarchical relationship formed after all actual business activities of the target enterprise are decomposed; Step S2 is specifically as follows: Step S2.1, inputting each actual business activity of the target enterprise into the unified enterprise model UEM; Step S2.2, using the business domain model of the unified enterprise model UEM, analyzing the actual business activities, and dividing the actual business activities into business domains of multiple target enterprises based on the rules of UEM; Step S2.3, input each of the business domains and the actual business activities into the operation process reference model; the operation process reference model decomposes the actual business activities and the business activities planned to be carried out in the future from the operation perspective, and obtains the target enterprise operation process and operation activities with a hierarchical structure of business domain-operation process-operation activity, which is the operation activity decomposition framework; Step S2.4, inputting each of the business domains and the actual business activities into a management element reference model; The management element reference model decomposes the actual business activities from a management perspective to obtain the target enterprise management theme, management goal and management elements in a hierarchical structure of business domain-management theme-management goal-management element, which is the management activity decomposition framework; Step S3, comprehensively analyzing the target enterprise's operation activity decomposition structure and management activity decomposition structure, and obtaining the operation activity required for data analysis of each management element in the business domain, and the operation activity is called the operation process data node BPN of the management element; Construct the data analysis requirements of each corresponding operation process data node BPN of the management element, and obtain the BPN required by the management element for data analysis, the required data analysis requirement table DN, the data encryption policy table DEP and the privacy data processing policy table PDP; Step S4, summarizing the data analysis requirements of all business domains of the target enterprise according to the operation process data node BPN, obtaining all data analysis requirements for each operation process data node BPN, and forming a data analysis requirement table DN, a data encryption policy table DEP, and a privacy data processing policy table PDP belonging to the operation process data node BPN; Step S5: For each operation process data node BPN, a detailed business data table DBDT unique to the operation subject is established: For each job process data node BPN, multiple job topics of the next level contained in the job process data node BPN are generated according to the management attributes; for each job topic, its data analysis requirements are associated, and the required data analysis requirement table DN, data encryption policy table DEP and privacy data processing policy table PDP are constructed; Construct a detailed business data table DBDT uniquely corresponding to each job subject, so that the detailed business data table DBDT set formed by all job subjects under the same job process data node BPN at least includes all data elements in the data analysis requirement table DN under the job process data node BPN, and all policies in the data encryption policy table DEP and the privacy data processing policy table PDP under the job process data node BPN; Step S6, establishing a dynamic DBDT management directory; storing the DBDT management directory and all detailed business data tables DBDT of the target enterprise into the DWR data report layer; At this point, before developing for data analysis needs, the enterprise data platform EDP including the target enterprise model EM of the target enterprise and the DWR data report layer of all detailed business data tables DBDT is pre-built; The enterprise data platform EDP is used to develop data analysis services for various data analysis needs from various business domains.

2. The method for constructing an enterprise data platform according to claim 1, characterized in that: Step S1 is specifically as follows: The unified enterprise model UEM includes a business domain model, an operation process reference model, a management element reference model and a business domain-management subject-operation process comparison table; The business domain model is used to define the rules for dividing the business activities of an enterprise into multiple business domains, thereby dividing the business activities of an enterprise into a number of business domains that are independent but interconnected, classified, and have a mutual support and driving relationship, and describes the superior-subordinate relationship between each of the business domains. The divided business domains are applicable to various types of target enterprises; The operation process reference model is used to define the rules for decomposing the business activities of an enterprise from the operation perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-operation process-operation activity". Therefore, for the business activities of each business domain, the operation process reference model is used to decompose the corresponding operation process, and then the operation process is decomposed into multiple operation activities. The decomposed operation processes and operation activities include the operation processes and operation activities of various types of target enterprises. The management element reference model is used to define the rules for decomposing the business activities of an enterprise from a management perspective, and decomposes the business activities of an enterprise according to the hierarchical structure of "business domain-management theme-management goal-management element". Therefore, for the business activities of each business domain, the management goal under the set management theme is further decomposed into multiple management elements, that is, the management elements that have a direct impact and driving relationship with the corresponding management theme and management goal. The decomposed management themes, management goals and management elements include the management themes, management goals and management elements of various types of target enterprises. The business domain-management subject-operation process comparison table is used to describe the association relationship between the operation process reference model and the management element reference model. Specifically, in the business domain, the management subject and the operation process are matched, and then the management element of the next level of the management target of the next level of the management subject and the operation activity of the next level of the operation process are established.

3. The method for constructing an enterprise data platform according to claim 1, characterized in that: Step S3 is specifically as follows: Step S3.1, based on the business domain-management theme-operation process comparison table, comprehensively analyze the operation activity decomposition structure and management activity decomposition structure of the target enterprise, and obtain the operation activity that needs to provide data for data analysis of each management element in the business domain, that is, the operation process data node BPN of the management element; Step S3.2, construct the data analysis requirement table DN: Each operation activity generates multiple data objects; after a data object is structured, it is a data table, and the data table has multiple data elements, and one data element is a field; the data table is stored in the enterprise data platform; For each management element, the constructed data analysis requirement table DN includes the following fields: business domain, management subject, management goal, management element, required BPN, required measurement data, and required dimension data; the required measurement data refers to the measurement data extracted from the data elements in the required data object; the required dimension data refers to the dimensions formed by some or all of the data elements in the required data object; Step S3.3, construct the data encryption policy table DEP: For each management element, a data encryption policy table DEP is constructed, including the following fields: business domain, management subject, management target, management element, required BPN, required data object, required data element, condition and encryption policy; its meaning is: when the value of a data object or a data element of a certain BPN required by a certain management element meets a certain condition, the corresponding encryption policy needs to be adopted for the data; Step S3.4, construct the privacy data processing policy table PDP: For each management element, a privacy data processing policy table PDP is constructed, including the following fields: business domain, management subject, management objective, management element, required BPN, required data object, required data element, condition and privacy data processing strategy; its meaning is: when the value of a data object or a data element of a BPN required by a management element meets a certain condition, the corresponding privacy data processing strategy needs to be adopted for the data.

4. The method for constructing an enterprise data platform according to claim 3, characterized in that: In step S5, the detailed business data table DBDT uniquely corresponding to the operation subject includes a DBDT master table, a data encryption policy table DEP and a privacy data processing policy table PDP; The DBDT main table includes the following fields: Business domain, work process, work activity, work subject, measurement, modification and dimension, which means: for a work subject under the business domain, work process and work activity, the corresponding work data is stored in the form of measurement, modification and dimension; The data encryption policy table DEP and the privacy data processing policy table PDP are used to define the data encryption policy and privacy data processing policy to be adopted when the DBDT main table provides corresponding job data.

5. The method for constructing an enterprise data platform according to claim 3, characterized in that: The detailed business data table DBDT uniquely corresponding to the operation theme established in step S5 becomes a standardized and reusable DBDT because it contains data required for all possible analysis requirements of various management elements in all business domains of the target enterprise.

6. The method for constructing an enterprise data platform according to claim 3, characterized in that: Also includes: Step S7, performing unique management and maintenance on each detailed business data table DBDT based on the DBDT management directory; When it is necessary to adjust and modify the established detailed business data table DBDT, trace back and find the related indicators and data reports that have referenced the detailed business data table DBDT according to the data reference relationship, and modify the existing indicators and data reports according to the data logic of the modified detailed business data table DBDT.

7. An enterprise data platform constructed by the method for constructing an enterprise data platform according to any one of claims 1 to 6, characterized in that: include: Target enterprise model EM, ODS source data layer, DWI data integration layer, DWR data reporting layer and DM data mart layer; The target enterprise model EM is used to describe the rules and results of decomposition of the target enterprise's business activities, thereby decomposing the enterprise's business activities into interrelated elements at different levels from different perspectives; The ODS source data layer is used to collect and store various types of original business activity data generated by enterprise business activities in an event-driven manner; The DWI data integration layer is used to clean and integrate the original business activity data of the ODS source data layer, and then obtain a list of operating activities matching the target enterprise model EM according to the target enterprise model EM, form an operating process data node BPN, and store it; The DWR data report layer is used to establish and store a unique detailed business data table DBDT for multiple operation themes according to data analysis requirements; The DM data mart layer is used to construct a DM layer data model based on the detailed business data table DBDT unique to multiple operation subjects of the DWR data report layer; Based on the DM layer data model, the data mart DM data generated by indicator and dimension design is designed and developed to provide a data foundation for various data analyses of the enterprise.

Images