A multi-service unit-oriented data collaborative operation method

Abstract

This application provides a data collaborative operation method for multiple business units, relating to the field of data management and processing technology. The method includes: collecting raw data from heterogeneous data sources corresponding to multiple business units, and standardizing the raw data based on preset industry data standards to generate standardized data; constructing a data asset catalog integrating industry knowledge based on the standardized data, and proactively recommending at least one data service to the corresponding business unit according to at least one target business scenario through the data asset catalog; and constructing an operation platform that supports data analysis result sharing and collaboration between at least two business units based on the data asset catalog and the proactively recommended data services.

Description

Technical Field

[0001] This application belongs to the field of data management and processing technology, specifically relating to a data collaborative operation method for multiple business units. Background Technology

[0002] In today's big data era, enterprises or organizations typically have multiple independent business units, such as R&D, production, sales, and quality inspection. These business units often use different information systems during operation, resulting in data being scattered across multiple heterogeneous data sources with vastly different data formats, encoding rules, and update frequencies. Existing data governance or data platform solutions usually focus on the technical aspects of data integration and processing, but they suffer from the following significant drawbacks: Data fusion is difficult: Due to the lack of unified data standards that are deeply integrated with specific industries or businesses, raw data from different business units are difficult to effectively correlate and merge, forming "data silos". This makes global data analysis and cross-business traceability (such as quality traceability) extremely time-consuming and labor-intensive.

[0003] Passive and inefficient data services: Traditional data platforms often provide passive data query or reporting services, requiring business personnel to clearly understand the location and meaning of the data they need. Data assets lack intelligent organization and presentation centered on business scenarios, failing to proactively push suitable data or services to the business units that need them, resulting in insufficient release of data value.

[0004] Lack of collaborative operation mechanisms: Although simple data sharing methods exist, there is a lack of an operational platform to support continuous and in-depth collaboration. The data analysis results and knowledge generated by various business units are difficult to share, reuse, and collaboratively develop within the organization securely and conveniently, resulting in repetitive work, long innovation cycles, and insufficient accumulation of knowledge assets. Summary of the Invention

[0005] This application provides a data collaborative operation method for multiple business units to solve one of the aforementioned technical problems.

[0006] The technical solution adopted in this application is as follows: This application provides a data collaborative operation method for multiple business units, including: S1. Collect raw data from heterogeneous data sources corresponding to multiple business units, and standardize the raw data based on preset industry data standards to generate standardized data; S2. Based on the standardized data, construct a data asset catalog that integrates industry knowledge, and proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario; S3. Based on the data asset catalog and proactively recommended data services, construct an operation platform that supports data analysis results sharing and collaboration between at least two business units.

[0007] According to one embodiment of this application, in step S1, the standardization processing of the raw data based on a preset industry data standard includes: According to industry production standards, the different codes representing the same entity in the original data are mapped and converted. And / or, align data with inconsistent time granularity in the original data.

[0008] According to one embodiment of this application, step S1, which involves collecting raw data from heterogeneous data sources corresponding to multiple business units, includes: Data is collected by deploying a dedicated data acquisition adapter corresponding to a specific business unit. The dedicated data acquisition adapter is used to collect the unique parameters of that business unit.

[0009] According to one embodiment of this application, in step S2, constructing a data asset catalog that integrates industry knowledge includes: The standardized data is scanned using a built-in industry keyword library to identify and tag data assets related to industry knowledge. Identified data assets are classified and managed according to their update frequency and / or sensitivity level.

[0010] According to one embodiment of this application, in step S2, the step of proactively recommending at least one data service to the corresponding business unit through the data asset catalog based on at least one target business scenario includes: In response to the detection of a triggering event for a target business scenario, data assets associated with the triggering event are filtered from the data asset catalog; The selected data assets and / or data service interfaces encapsulated based on the data assets are pushed to the business unit that processes the target business scenario.

[0011] According to one embodiment of this application, in step S3, the construction of an operation platform that supports data analysis result sharing and collaboration between at least two business units includes: Establish dedicated data marts or analytical model libraries for specific technical fields, enabling authorized business units to conduct collaborative modeling and analysis without leaving their local systems.

[0012] According to one embodiment of this application, the method further includes: Build a knowledge base for associated defect patterns, root cause analysis, and improvement measures; In response to the detection of a new defect pattern in the operating platform, the system automatically matches and pushes the corresponding root cause analysis and improvement measures that have been recorded in the history of the knowledge base.

[0013] According to one embodiment of this application, the method further includes a quality control step: The standardized data generated in step S1 is verified based on preset industry-specific quality rules, which include quantitative indicators for data integrity, accuracy, and / or timeliness.

[0014] According to one embodiment of this application, the method further includes a security control step: For data assets marked as sensitive in the data asset catalog, implement field-level encrypted storage and transmission control; And / or, automatically add traceable digital watermarks to the data output exported from the operating platform.

[0015] A second aspect of this application provides a data collaborative operation system for multiple business units, including: The data acquisition and standardization module is used to collect raw data from heterogeneous data sources corresponding to multiple business units, and to standardize the raw data based on preset industry data standards to generate standardized data. The data asset and service module is communicatively connected to the data acquisition and standardization module. It is used to construct a data asset catalog that integrates industry knowledge based on the standardized data, and to proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario. The collaborative operation platform module communicates with the data asset and service module and is used to build an operation platform that supports the sharing and collaboration of data analysis results between at least two business units based on the data asset catalog and proactively recommended data services.

[0016] Due to the adoption of the above technical solution, the beneficial effects achieved by this application are as follows: This application transforms and aligns raw data from different sources and in different formats according to industry data standards closely integrated with business operations, generating standardized data with consistent semantics and standardized formats. This lays a reliable technical foundation for subsequent cross-business unit data correlation analysis and unified management, fundamentally eliminating "data silos".

[0017] The standardized processing itself includes basic checks on data integrity and accuracy, making the quality of the generated datasets controllable and directly usable for advanced analysis and service encapsulation, thus reducing the time and cost of data preprocessing.

[0018] The constructed data asset catalog is not simply a list of metadata, but incorporates industry-specific knowledge (such as business terminology, process parameters, and compliance requirements), enabling data resources to be organized, classified, and retrieved in a way that is easy for business personnel to understand, greatly improving the visibility and understandability of data assets.

[0019] By identifying target business scenarios (such as new product development and quality anomaly investigation), the system can proactively match and push relevant data services (such as specific analysis reports, API interfaces, and datasets) from the catalog, transforming "people searching for data" into "data finding people." This significantly shortens the path for business personnel to obtain key information and tools, and improves the speed and accuracy of decision-making response.

[0020] The operations platform provides a secure and convenient environment for sharing, reusing, and collaboratively developing data analysis results (such as models, reports, and optimization solutions) across different business units. This promotes knowledge flow and experience accumulation within the organization.

[0021] Based on the platform, the analytical results of one business unit can be quickly verified, applied, or iterated upon by other units, forming a virtuous cycle of "data generating knowledge, knowledge empowering business, and business feeding back data." This not only reduces the cost of redundant R&D but also accelerates the overall process of data-driven business innovation and optimization. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating a data collaborative operation method for multiple business units, provided as an embodiment of this application. Detailed Implementation

[0023] To more clearly illustrate the overall concept of this application, a detailed explanation is provided below with reference to the accompanying drawings.

[0024] Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of this application is not limited to the specific embodiments disclosed below. It should be noted that, unless otherwise specified, the embodiments of this application and the features thereof can be combined with each other.

[0025] In this application, unless otherwise expressly specified and limited, the "above" or "below" of the second feature can mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. In the description of this specification, references to terms such as "an embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples.

[0026] Example 1 like Figure 1 As shown, a data collaborative operation method for multiple business units includes: S1. Collect raw data from heterogeneous data sources corresponding to multiple business units, and standardize the raw data based on preset industry data standards to generate standardized data.

[0027] Specifically, raw data is collected from heterogeneous data sources corresponding to multiple business units. In practice, this is achieved by deploying dedicated data acquisition adapters. For the PDM system, deploy adapter A, which is dedicated to collecting unique parameters such as design version and material specifications of key components.

[0028] For the MES system, deploy adapter B, which is dedicated to collecting production-specific parameters such as "spindle vibration spectrum" and "assembly torque value".

[0029] For CRM and operations and maintenance platforms, deploy adapter C, which is specifically used to collect after-sales specific parameters such as "fault codes" and "on-site environmental temperature and humidity".

[0030] Then, the raw data is standardized based on preset industry data standards. The company has formulated and preset the "High-end Equipment Manufacturing Data Element Specification" as the industry data standard. The processing includes: Mapping and Conversion: Based on industry production standards (such as ISO standard part codes), the internal "part drawing numbers" in PDM, the "workshop material codes" in MES, and the "maintenance spare parts SKUs" in CRM are uniformly mapped and converted to the standard "globally unique material identifier".

[0031] Alignment Processing: Align the device status data collected every second in the MES with the product operation status data reported every 5 minutes by the after-sales platform in terms of time granularity. For example, aggregate high-frequency device data in 5-minute windows (calculate the average and maximum values) so that it can be correlated and analyzed with after-sales data on the same time scale.

[0032] After the above processing, standardized data with consistent semantics across design, production, and after-sales is generated and stored in the enterprise data lake.

[0033] S2. Based on the standardized data, construct a data asset catalog that integrates industry knowledge, and proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario.

[0034] Specifically, construct a data asset catalog that integrates industry knowledge: The system scans the standardized data in the data lake using a built-in industry keyword library (which includes terms such as "fatigue life", "tolerance fit", and "early warning threshold").

[0035] Identify and label data assets closely related to industry knowledge, such as "core component stress simulation dataset", "historical assembly quality deviation report", and "typical failure mode spectrum library".

[0036] Data assets are categorized and managed according to their update frequency and / or sensitivity level: for example, real-time updated "production line OEE (Global Equipment Efficiency) dashboard data" are marked as "high-frequency dynamic data"; "core transmission component design drawing set" are marked as "high-sensitivity data" to trigger a higher-level access approval process.

[0037] Based on the target business scenario, proactively recommend data services: Suppose a target business scenario occurs: The manufacturing department reports in the MES that "a batch of products has a coaxiality error alarm on the assembly line".

[0038] Triggering event: This alarm event is detected by the system as a triggering event for a "quality anomaly scenario".

[0039] Filter related assets: The system immediately and automatically filters out the "Design BOM and tolerance requirements", "Incoming material inspection reports of the parts used in this batch", "Historical assembly process parameter packages of similar products", and "Root cause analysis cases of past coaxiality problems" from the data asset catalog.

[0040] Push service: The system will push the data assets selected above, along with a pre-packaged "assembly process parameter optimization suggestion data service interface", to the homepage of the office portal of the process engineer in the production and manufacturing department.

[0041] S3. Based on the data asset catalog and proactively recommended data services, construct an operation platform that supports data analysis results sharing and collaboration between at least two business units.

[0042] Specifically, based on the data asset catalog and proactively recommended data services, the company has built an internal innovation and operation platform.

[0043] Establish a dedicated data mart or analysis model library: For example, the design and R&D department can publish the "lightweight design optimization model based on finite element analysis" accumulated over many years to the platform's "advanced design model library".

[0044] Support for cross-unit collaborative modeling and analysis: When analyzing a type of high-frequency fault, the after-sales service department discovered that it might be related to early design. Through the platform, after-sales engineers can request collaborative analysis from the R&D department without the data leaving their local database (after-sales database). The R&D department's optimization model is authorized to run on the after-sales data in the platform's sandbox environment, jointly locating the problem of insufficient design margin, while neither party's original sensitive data is directly exposed.

[0045] Building and applying a knowledge base: The platform has built a "failure mode-root cause-improvement measure knowledge base". When the manufacturing department enters a new "abnormal vibration mode" identified by the sensor on the platform, the system responds by automatically matching and pushing "bearing preload adjustment" improvement measures that have been jointly verified by the R&D and after-sales departments in the past from the knowledge base.

[0046] According to one embodiment of this application, in step S1, the standardization processing of the raw data based on a preset industry data standard includes: According to industry production standards, the different codes representing the same entity in the original data are mapped and converted. And / or, align data with inconsistent time granularity in the original data.

[0047] Specifically, the first step is to map and convert different codes representing the same entity in the original data according to industry production standards. In actual business operations, the same entity (such as a specific material, a production device, or a process step) may be identified by completely different internal codes in different business unit systems. For example, the material warehouse in the R&D system may use a classification-based code "MAT-Design-001," while the production execution system may use a serial number-based code "P10001," and the warehousing system may use a code conforming to the industry standard "GB / T-XXXXX." This inconsistency makes it impossible to directly associate the data. This step establishes a mapping table or conversion rules between different coding systems by pre-establishing or referencing unified industry production standards (such as national material coding standards or enterprise master data standards). During processing, the system automatically converts different codes from various sources that point to the same entity into a unified, standardized identifier based on this mapping relationship, thereby ensuring that the entity can be uniquely and consistently identified and associated in all subsequent processing.

[0048] Secondly, the process involves aligning data with inconsistent time granularities in the original data. Different data sources often have different time granularities (i.e., the frequency and cycle of data generation or recording) due to differences in business characteristics and technical architecture. For example, a sensor might collect temperature data once per second (second-level granularity), a production report might be recorded once per work order completed (hourly or daily granularity), while a quality inspection report might be generated once per batch (batch cycle, possibly several days). This difference in time granularity makes it impossible to directly compare or fuse different data along the time dimension. This step addresses this issue through time alignment. Specific alignment methods can include aggregation and correlation: for high-frequency data (such as sensor data), aggregation operations (such as averaging or taking the maximum value) can be performed according to the cycle required for business analysis (such as per hour, per batch); for low-frequency data, a clear timestamp or time interval is established, and it is correlated with the aggregated high-frequency data on a unified time axis. Through this process, data with inconsistent time scales are transformed to the same or comparable time benchmark, laying the foundation for cross-business time-series analysis and event correlation.

[0049] According to one embodiment of this application, step S1, which involves collecting raw data from heterogeneous data sources corresponding to multiple business units, includes: Data is collected by deploying a dedicated data acquisition adapter corresponding to a specific business unit. The dedicated data acquisition adapter is used to collect the unique parameters of that business unit.

[0050] Specifically, in practical applications, the information systems used by different business units within an enterprise (such as R&D, production, and supply chain) often differ significantly in their technical architecture, data storage formats, interface protocols, and security policies, constituting heterogeneous data sources. General-purpose data acquisition tools or standard interfaces often cannot be directly adapted to all systems, and it is particularly difficult to acquire core business data in various business scenarios.

[0051] Therefore, this step employs a customized deployment of dedicated data acquisition adapters. Each adapter is developed or configured for a specific information system used by a particular business unit (e.g., a Manufacturing Execution System (MES) for a production unit, or a Product Lifecycle Management (PLM) system for a research and development unit). This adapter has a deep understanding of the target system's data model, access interfaces (such as direct database connections, API calls, and file parsing), and communication protocols, enabling it to reliably and efficiently extract raw data from this heterogeneous data source.

[0052] Crucially, one of the core functions of the dedicated data acquisition adapter is to collect parameters specific to each business unit. This means that the adapter does not merely acquire general information, but can identify and extract specific data fields that reflect the core business status, process, or professional field of that unit. For example, the adapter for the production unit will specifically collect production-specific parameters such as "equipment OEE (Global Equipment Efficiency)," "process cycle time," and "work-in-process quantity"; while the adapter for the quality unit will specifically collect quality-specific parameters such as "inspection batch number," "defect code," and "measurement deviation value." In this way, it is ensured that the data sets collected from each business unit can completely and accurately reflect the business essence of its professional field, providing a high-quality data foundation for subsequent cross-domain data fusion and collaborative analysis.

[0053] According to one embodiment of this application, in step S2, constructing a data asset catalog that integrates industry knowledge includes: The standardized data is scanned using a built-in industry keyword library to identify and tag data assets related to industry knowledge. Identified data assets are classified and managed according to their update frequency and / or sensitivity level.

[0054] Specifically, the first step involves scanning the standardized data using a built-in industry keyword library to identify and label data assets related to industry knowledge. This requires pre-constructing an industry keyword library, which includes specialized terminology, core business concepts, key process parameters, and compliance standards within a specific industry sector. The system utilizes this keyword library to automatically scan and semantically analyze the standardized data processed in step S1. When a data table, field name, data content, or its description contains or matches terms from the keyword library, the system identifies that portion of the data as a data asset with industry value. Subsequently, the system labels these assets with corresponding industry-specific tags or classification identifiers. For example, data tables containing "yield strength" or "heat treatment curve" are labeled with the "materials and processes" theme, and data involving "defect rate" or "SPC control chart" are labeled with the "quality analysis" theme. This process elevates the raw data into assets with clear industry semantic annotations.

[0055] Secondly, the identified data assets are classified and managed according to their update frequency and / or sensitivity level. After identification and labeling, the system further automates the classification of data assets based on management dimensions.

[0056] Data assets are categorized by update frequency: The system classifies data assets based on the time period during which they are written to or updated. For example, sensor data streams updated every second from the production site are categorized as "real-time high-frequency assets"; daily production reports are categorized as "daily batch assets"; and relatively stable product information is categorized as "static reference assets." Assets with different update frequencies will be adapted to different storage, computing, and quality of service (QoS) strategies.

[0057] Data assets are categorized by sensitivity level: The system determines the sensitivity of data assets based on preset security rules or content recognition models. For example, assets containing core formulas, personally identifiable information, or undisclosed financial data are marked as "highly sensitive assets"; general internal operational data are marked as "internal assets"; and publicly available industry reports are marked as "public assets." Different access control, encryption, and auditing policies will be implemented for assets with different sensitivity levels.

[0058] Through the above scanning, identification, labeling, and classification management, a structured data asset catalog that not only lists data resources but also deeply integrates industry semantic understanding and management strategies is finally constructed, laying the foundation for subsequent scenario-based intelligent service recommendations.

[0059] According to one embodiment of this application, in step S2, the step of proactively recommending at least one data service to the corresponding business unit through the data asset catalog based on at least one target business scenario includes: In response to the detection of a triggering event for a target business scenario, data assets associated with the triggering event are filtered from the data asset catalog; The selected data assets and / or data service interfaces encapsulated based on the data assets are pushed to the business unit that processes the target business scenario.

[0060] Specifically, firstly, the system needs to respond to the detection of triggering events for the target business scenario. This means that the system monitors or receives specific signals from various business systems in real time, indicating that a predefined target business scenario has occurred or is about to require processing. The triggering event can be a specific business operation (e.g., a user creating an "abnormal product handling order" in the quality management system), an automated system alarm (e.g., a "critical parameter exceeding limit" alarm from production equipment), or a periodic task node (e.g., the initiation of the "monthly business analysis report" generation task). The system captures these events by listening to interfaces, parsing messages, or polling status.

[0061] Upon detecting a triggering event, the system immediately filters data assets associated with the triggering event from the data asset catalog. This process is based on association rules or semantic matching models between events and data assets. The system analyzes the attributes of the triggering event (such as event type, involved entities, occurrence time, and key parameters) and matches them with industry theme tags, metadata descriptions, and content keywords of each asset in the data asset catalog. For example, when the triggering event is "a foreign object complaint in a batch of products," the system will automatically filter out a series of data assets that are strongly related to the event in terms of business logic, such as "production records of this batch of products," "suppliers and inspection reports of the raw materials used," "cleaning and disinfection logs of the production line during the same period," and "investigation reports of similar complaints in the past."

[0062] Finally, the system performs a recommendation action, which involves pushing the selected data assets and / or data service interfaces encapsulated based on those data assets to the business units processing the target business scenario. Specifically: Pushing data assets themselves: The system can directly send access links and summary information of the filtered data tables, documents or datasets to business personnel or relevant application systems handling the scenario.

[0063] Pushing encapsulated data service interfaces: Going further, the system can call services that have been developed in advance based on relevant data assets and registered to the platform, such as an API interface called "Quality Traceability Analysis" or an executable analysis model called "Comparison of Similar Defect Cases", and push the entry point or result of calling the service directly to the user.

[0064] The goal of the push notifications is to ensure that information is accurately delivered to the business units handling the target business scenarios. For example, quality-related data and tools are pushed to the quality management department, and production-related optimization suggestions are pushed to the production and manufacturing units. Push channels can include internal communication tools, business system workbenches, dedicated portals, or email.

[0065] Through the closed-loop process of "event perception - intelligent matching - targeted push" described above, this invention achieves a fundamental transformation from passive querying to proactive, accurate, and scenario-based data service provision.

[0066] According to one embodiment of this application, in step S3, the construction of an operation platform that supports data analysis result sharing and collaboration between at least two business units includes: Establish dedicated data marts or analytical model libraries for specific technical fields, enabling authorized business units to conduct collaborative modeling and analysis without leaving their local systems.

[0067] Specifically, dedicated data marts or analytical model repositories are established: the operations platform will build one or more centralized storage and publishing centers, namely "dedicated data marts" or "analytical model repositories." These marts or model repositories are not general-purpose repositories, but are organized and built for specific technical fields. For example, in manufacturing companies, a "predictive maintenance algorithm model repository" might be established, specifically collecting various algorithms related to equipment failure prediction; or in pharmaceutical companies, a "clinical trial data mart" might be established, specifically storing standardized clinical trial data. The datasets or models in these repositories are usually validated and valuable data analysis results formed by various business units in their past practices and innovations.

[0068] Authorized Access Mechanism: The platform implements a strict access control system. Only authorized business units can be granted access, use, and even contribution permissions to specific data marts or model libraries based on their roles, projects, or collaboration needs. This ensures the secure and controlled sharing of knowledge assets.

[0069] Supporting collaborative modeling and analysis without data leaving local storage: This is a key technical feature for achieving secure collaboration. The platform provides a computing framework or environment that allows a business unit (such as a research and development center) to deploy or provide API calls to its self-developed analytical models (such as a quality prediction algorithm) that have been published to the model library. When another authorized business unit (such as a production plant) needs to use the model to perform computational analysis on its local data, it does not need to transfer its sensitive raw data out of the local storage environment. Instead, the computing framework can employ technologies such as federated learning, secure multi-party computation, or model API sandbox calls. Locally at the production plant, the model performs computations on local data in a controlled environment, returning only the analysis results (such as prediction conclusions and model gradient update parameters) that do not contain details of the raw data to the platform or model provider, thereby enabling cross-unit joint analysis or iterative model optimization. This approach effectively promotes the collaboration and value creation of knowledge (models) and data across different business units while fully protecting the data sovereignty and privacy of each unit.

[0070] According to one embodiment of this application, the method further includes: Build a knowledge base for associated defect patterns, root cause analysis, and improvement measures; In response to the detection of a new defect pattern in the operating platform, the system automatically matches and pushes the corresponding root cause analysis and improvement measures that have been recorded in the history of the knowledge base.

[0071] Specifically, a knowledge base is constructed to associate defect patterns, root cause analysis, and improvement measures: First, a structured knowledge base is built within the aforementioned operation platform. This knowledge base is used to systematically store and associate three types of core information: Defect patterns: Characterize and classify various anomalies, faults or non-conformities that occur during production, operation or service, such as "welding porosity", "software response timeout", "assembly dimension deviation", etc., and extract their key feature vectors.

[0072] Root cause analysis: For each recorded defect pattern, store the verified root cause analysis conclusions that led to the defect, such as "welding parameter current setting is too low", "database index is missing", "fixture positioning pin is worn".

[0073] Improvement measures: For each root cause analysis, a list of corrective and preventative measures that have proven effective in practice is stored, such as "adjust the welding current to XX-YY amperes", "add a composite index to a specific query field", and "inspect and replace locating pins on a planned basis for each batch".

[0074] This information is interconnected in a structured manner (such as in the form of triples, case records, etc.), forming a knowledge network that can be retrieved and reasoned about by computers.

[0075] In response to the detection of a new defect pattern in the operation platform, the system automatically matches and pushes historically recorded root cause analyses and improvement measures from the knowledge base: When the operation platform detects a new defect pattern (i.e., a currently occurring, characteristic anomaly) through data monitoring, user reporting, or analysis models, the system automatically triggers the following response process: Automatic matching: The system calculates and matches the feature descriptions or feature vectors of new defect patterns with numerous historical defect patterns in the knowledge base. The matching algorithm can be based on keywords, feature vector distance, or graph reasoning.

[0076] Push historical solutions: If one or more historically recorded defect patterns with similarity exceeding a preset threshold are found in the knowledge base, the system will automatically obtain the root cause analysis and improvement measures corresponding to the historical defect pattern, and generate notifications or reports for these historical solutions.

[0077] Targeted push: Finally, the system pushes this information, which includes historical root causes and measures, to the relevant business personnel or responsible systems handling the current new defects, providing them with direct decision support and clues for problem solving, thereby realizing the automated reuse of experience and knowledge and accelerating the problem investigation and resolution process.

[0078] According to one embodiment of this application, the method further includes a quality control step: The standardized data generated in step S1 is verified based on preset industry-specific quality rules, which include quantitative indicators for data integrity, accuracy, and / or timeliness.

[0079] Specifically, this step aims to perform automated quality verification on the standardized data generated in step S1. Its core is to perform the verification based on preset industry-specific quality rules.

[0080] The industry-specific quality rules are a set of quantifiable and executable inspection rules predefined based on the business specifications, management requirements, and technical standards of a specific industry. These rules primarily set clear quantitative indicator requirements for the following core quality dimensions of the data: Integrity verification: Rules set tolerance thresholds for the degree of missing information in essential fields or key information sets of data records. For example, a rule might require that the missing rate of the three fields "Production Date," "Supplier Code," and "Key Component Content" in a "Raw Material Batch Testing Report" must not exceed 1%. The system scans standardized data, calculates the proportion of null or invalid values ​​in the specified fields, and determines whether this quantitative indicator is met.

[0081] Accuracy verification: Rules set accuracy standards by comparing data with authoritative reference sources or verifying whether their value range and logical relationships conform to common business sense. For example, a rule might require that the deviation between the reading of "production process temperature" and the recorded value of a periodically calibrated reference instrument not exceed ±2℃; or it might require that the value of "product qualification rate" must be within a reasonable range of 0% to 100%. The system performs comparisons or logical checks to verify whether the data values ​​meet these accuracy constraints.

[0082] Timeliness verification: The rules set an upper limit on the overall time delay from the actual occurrence of data to its collection, processing, and generation into standardized data by the system. For example, the rules may require that the time delay of real-time status data generated by production line sensors from its generation time to its availability in the standardized data pool must be less than 5 seconds. The system calculates the delay and determines whether a timeout has occurred by comparing the data's timestamp with the current system time.

[0083] During the quality control process, the system automatically invokes these rules to scan and calculate standardized data streams or batches of data. Verification results are typically categorized as acceptable, warning, or unacceptable. For data that violates quantitative indicators, the system can execute predefined handling strategies, such as marking it with a quality flag, triggering an alarm notification, blocking its entry into subsequent stages, or initiating a data cleaning task, thereby ensuring collaborative operations are based on high-quality data.

[0084] According to one embodiment of this application, the method further includes a security control step: For data assets marked as sensitive in the data asset catalog, implement field-level encrypted storage and transmission control; And / or, automatically add traceable digital watermarks to the data output exported from the operating platform.

[0085] Specifically, field-level encryption and transmission control for sensitive data assets: This measure targets data assets that have been marked as sensitive in the data asset catalog (such as tables or fields containing core recipes, personal privacy information, or undisclosed financial data). Its implementation includes: Field-level encrypted storage: The system identifies specific sensitive fields in sensitive assets (such as "customer ID number" and "core additive ratio") and only stores the content of these fields after encrypting them with a strong encryption algorithm. Non-sensitive fields (such as "record generation time") are still stored in plaintext, thus achieving a balance between security and processing efficiency.

[0086] Transmission control: When these sensitive data assets or their encrypted fields need to be transmitted over the network (e.g., retrieved from a storage server to an application server for computation, or exchanged between different security domains), the system ensures that they remain encrypted during transmission (e.g., transmitted via a TLS-encrypted link). They are only decrypted and used on authorized endpoints or in secure computing environments with strict identity and authorization verification.

[0087] Add a source watermark to the exported data: This measure targets data outputs exported from the aforementioned operating platform (such as generated reports, downloaded datasets, and shared analytical charts). Its implementation is as follows: Automatic addition: When a user or system performs an export operation, the platform automatically triggers the watermark embedding process without manual intervention.

[0088] Add a traceable digital watermark: The system invisibly embeds a digital watermark containing traceable information into the exported file or data stream. This watermark information is typically uniquely associated with the current export operation and may include, but is not limited to: the exporter's identity, export timestamp, authorized session ID, and the recipient's identifier for the exported file. This watermark is concealed and robust, making it difficult to remove or tamper with using conventional methods.

[0089] Source tracing function: If the exported data is leaked without authorization, the digital watermark information can be extracted through professional detection tools or algorithms, thereby accurately tracing the source of the export operation (who and when), providing direct technical evidence for security incident investigation and responsibility determination.

[0090] By combining the above measures, a comprehensive security protection system covering static storage, dynamic transmission, and external circulation can be built while ensuring data availability and flow.

[0091] A second aspect of this application provides a data collaborative operation system for multiple business units, including: The data acquisition and standardization module is used to collect raw data from heterogeneous data sources corresponding to multiple business units, and to standardize the raw data based on preset industry data standards to generate standardized data. The data asset and service module is communicatively connected to the data acquisition and standardization module. It is used to construct a data asset catalog that integrates industry knowledge based on the standardized data, and to proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario. The collaborative operation platform module communicates with the data asset and service module and is used to build an operation platform that supports the sharing and collaboration of data analysis results between at least two business units based on the data asset catalog and proactively recommended data services.

[0092] For any parts not mentioned in this application, existing technologies may be used or referenced.

[0093] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0094] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A data collaborative operation method for multiple business units, characterized in that, include: S1. Collect raw data from heterogeneous data sources corresponding to multiple business units, and standardize the raw data based on preset industry data standards to generate standardized data; S2. Based on the standardized data, construct a data asset catalog that integrates industry knowledge, and proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario; S3. Based on the data asset catalog and proactively recommended data services, construct an operation platform that supports data analysis results sharing and collaboration between at least two business units.

2. The method according to claim 1, characterized in that, In step S1, the standardization process of the raw data based on preset industry data standards includes: According to industry production standards, the different codes representing the same entity in the original data are mapped and converted. And / or, align data with inconsistent time granularity in the original data.

3. The method according to claim 1 or 2, characterized in that, In step S1, collecting raw data from heterogeneous data sources corresponding to multiple business units includes: Data is collected by deploying a dedicated data acquisition adapter corresponding to a specific business unit. The dedicated data acquisition adapter is used to collect the unique parameters of that business unit.

4. The method according to claim 1, characterized in that, In step S2, the construction of a data asset catalog integrating industry knowledge includes: The standardized data is scanned using a built-in industry keyword library to identify and tag data assets related to industry knowledge. Identified data assets are classified and managed according to their update frequency and / or sensitivity level.

5. The method according to claim 1, characterized in that, In step S2, the step of proactively recommending at least one data service to the corresponding business unit through the data asset catalog based on at least one target business scenario includes: In response to the detection of a triggering event for a target business scenario, data assets associated with the triggering event are filtered from the data asset catalog; The selected data assets and / or data service interfaces encapsulated based on the data assets are pushed to the business unit that processes the target business scenario.

6. The method according to claim 1, characterized in that, In step S3, the construction of an operation platform that supports data analysis results sharing and collaboration between at least two business units includes: Establish dedicated data marts or analytical model libraries for specific technical fields, enabling authorized business units to conduct collaborative modeling and analysis without leaving their local systems.

7. The method according to claim 6, characterized in that, The method further includes: Build a knowledge base for associated defect patterns, root cause analysis, and improvement measures; In response to the detection of a new defect pattern in the operating platform, the system automatically matches and pushes the corresponding root cause analysis and improvement measures that have been recorded in the history of the knowledge base.

8. The method according to claim 1, characterized in that, The method also includes a quality control step: The standardized data generated in step S1 is verified based on preset industry-specific quality rules, which include quantitative indicators for data integrity, accuracy, and / or timeliness.

9. The method according to claim 1, characterized in that, The method also includes a security control step: For data assets marked as sensitive in the data asset catalog, implement field-level encrypted storage and transmission control; And / or, automatically add traceable digital watermarks to the data output exported from the operating platform.

10. A data collaborative operation system for multiple business units, characterized in that, include: The data acquisition and standardization module is used to collect raw data from heterogeneous data sources corresponding to multiple business units, and to standardize the raw data based on preset industry data standards to generate standardized data. The data asset and service module is communicatively connected to the data acquisition and standardization module. It is used to construct a data asset catalog that integrates industry knowledge based on the standardized data, and to proactively recommend at least one data service to the corresponding business unit through the data asset catalog according to at least one target business scenario. The collaborative operation platform module communicates with the data asset and service module and is used to build an operation platform that supports the sharing and collaboration of data analysis results between at least two business units based on the data asset catalog and proactively recommended data services.

Images