Data Center-Based Commissioning and Traceability Management System and Method
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA ELECTRONIC TECH GRP CORP NO 38 RES INST
- Filing Date
- 2026-02-04
- Publication Date
- 2026-05-26
Smart Images
Figure CN122086980A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data traceability technology for electronic equipment products, specifically to a data center-based commissioning and traceability management system and method. Background Technology
[0002] Currently, data management in the product development process of electronic equipment is mostly scattered across various business systems or relies on manual entry. When quality issues arise and product data traceability is required, the results and efficiency of traceability are affected by the dispersed nature of multi-source data, manual input errors, and poor real-time performance.
[0003] Chinese Patent Publication No. CN115759834A discloses a product data traceability method and system based on an information system. It develops inter-system data traceability functions by managing data in various systems corresponding to a product according to the technical status management approach, ensuring that data at each stage can be effectively traced and improving the convenience of data traceability for various products.
[0004] The production and manufacturing of electronic equipment products are usually carried out by multiple production units or departments working together. The commissioning process is highly dependent on the data generated by the preceding parts production, component assembly and commissioning and final assembly. Simply summarizing the data from multiple systems without considering the inconsistencies in data formats and incomplete information between multiple systems, equipment and manually entered data, and without considering the timeliness of data, can easily lead to the break of the traceability chain and affect the efficiency and accuracy of traceability management. Summary of the Invention
[0005] The technical problem to be solved by this invention is how to overcome the problems caused by heterogeneous data formats of test equipment, inconsistent standards for recording key parameters, delayed updates of process data, and lack of effective correlation of multi-source heterogeneous data during the commissioning and testing of electronic equipment, such as broken traceability chains, distorted analysis and comparison, delayed decision response, and one-sided root cause localization, so as to achieve full-process, high-fidelity, and correlated digital traceability and in-depth analysis.
[0006] The data center-based commissioning and traceability management system includes the following modules:
[0007] The forward traceability module is used to model entities such as product structural units, material batches, test nodes, operators, and process versions as nodes, and to abstract the business relationships between entities, such as assembly, testing, and inspection, as directed edges with timestamps and relationship weights, forming a physical relationship graph of equipment commissioning with time sequence attributes; based on the physical relationship graph of equipment commissioning, the complete assembly and testing process from components to the whole machine can be traced through path search. The commissioning report association and retrieval module extracts key entities such as product serial number, test item, time, and commissioning equipment from various commissioning reports, transforms them into standardized metadata, and automatically matches and associates this standardized metadata with the structured product physical bill of materials, process route nodes, and commissioning tasks in the data center, thereby constructing a multi-dimensional index relationship between report, physical object, and process stage. Users can retrieve and trace the source using key fields such as product serial number, time range, and test type. The reverse tracing module is used to reverse analyze and calculate all potentially affected upper-level assemblies and final product lists based on the constructed equipment commissioning physical relationship diagram, combined with material loading relationships, test dependencies, and historical fault association rules. It also marks the propagation path and confidence level of the impact, and prioritizes the detection or disposal of upper-level assemblies with confidence levels higher than a preset threshold.
[0008] This invention forms a physical relationship diagram of equipment commissioning with time-series attributes, breaking through the limitations of traditional tree structures in expressing complex relationships across levels and batches. Based on the physical relationship diagram of equipment commissioning, the complete assembly and testing process from components to the whole machine can be traced through path search. In addition, the commissioning report association and retrieval module realizes automatic matching and association of metadata with the structured physical BOM of products, process route nodes and commissioning tasks in the data center. The reverse tracing module quickly locates the impact range of the problem source, providing quantitative basis for risk assessment and disposal decisions. The overall solution realizes full-process, high-fidelity, and associative digital traceability and in-depth analysis.
[0009] Furthermore, the forward tracing module is also used for: By analyzing various data from the design bill of materials, manufacturing bill of materials, and production execution process of electronic equipment products, a dynamic graph data model is constructed. This model models each uniquely identified physical unit as a graph node, assigning it attributes such as physical serial number, material code, and status. Simultaneously, each recorded assembly operation or test node activity is modeled as a directed edge, assigning attributes including operation type, timestamp, operation result, and the ID of the operator or equipment. A composite index is established for node and edge types. When a user queries the destination of a component corresponding to a specific node type, a depth- or breadth-order traversal is performed along the edges corresponding to that edge type. Irrelevant paths are filtered using the index, returning the complete assembly chain from the component to the final integrated machine, and visually displayed in the form of a topology graph.
[0010] Furthermore, the physical unit with independent identification includes an electronic equipment unit with a serial number, a transmitting or receiving module, a printed circuit board, and components; the operation type includes assembly and testing.
[0011] Furthermore, the node types include components, modules, and complete machines, and the edge types include assembly and testing.
[0012] Furthermore, the forward tracing module is also used for: Starting from the target physical node, a directed traversal is performed along predefined relational edges using a graph database to collect all test tasks, equipment, and debugging nodes associated with the target physical node, forming an original relational subgraph. Here, predefined relational edges refer to relational edges with the target physical node as the object under test. When tracing the testing history of a complete electronic equipment, all test tasks, equipment, and debugging nodes related to the complete electronic equipment are first extracted from the original relational subgraph. The timestamp attributes of the corresponding edges of the test tasks, equipment, and debugging nodes are then extracted and globally sorted as tracing tasks. Subsequently, the tracing tasks are grouped by day or by test stage. Finally, the results are presented through a timeline view, showing when the complete equipment underwent what kind of test, which equipment was used, and what the result was, realizing dynamic playback of the test history.
[0013] Furthermore, the commissioning report association and retrieval module is also used for: Configure the association rule base to define how to map report metadata to structured business objects in the data center. The mapping process is as follows: if the product serial number in the report metadata matches the serial number field of a physical node in the product's physical bill of materials exactly, and the report test time is between the planned start time and the actual completion time of a commissioning task in the process route for that node, then establish a bidirectional association between the report and the physical node and the commissioning task node.
[0014] Furthermore, the report metadata includes the product serial number, test item code, and timestamp.
[0015] Furthermore, the reverse tracing module is also used for: When a user selects a faulty component to initiate reverse tracing, a reverse propagation simulation analysis is performed starting from the faulty component in the equipment commissioning physical relationship diagram. The reverse propagation simulation analysis starts from the source node and traverses along the reverse edge. The confidence level of each parent node reached is calculated as follows: ,in, For the current node The impact confidence level parent node confidence level For the node To the parent node The propagation edge weights, The path decay factor, For from the source node to The number of edges, the comprehensive impact confidence of each parent node is the maximum value of the impact confidence calculated in all propagation paths; all propagation paths are recorded synchronously, and an intelligent analysis report is output at the end of the tracing. The intelligent analysis report includes an impact list of all affected upper-level products or modules sorted from high to low confidence, a propagation map of the impact diffusion path, and key recommendations for priority detection or disposal of upper-level assemblies with confidence levels higher than the preset threshold.
[0016] Furthermore, the propagation edge weights are obtained as follows: In the physical relationship graph of equipment commissioning, the weight of the edge is set to 1 for the influence that will inevitably propagate back to the parent node along the path. For the edge with test dependency, the weight is set according to the test dependency relationship. For example, if test A fails, test B has an x% probability of also failing, then the weight of the propagation edge between the two is 0.0x.
[0017] This invention also provides a method for a data center-based commissioning and traceability management system, comprising: S501: The data center connects to the business systems of all stages in the entire lifecycle of electronic equipment product development, including design, process, production, and commissioning. It achieves full collection of structured data and unstructured documents and records. The system uses the product's physical serial number as a unique identifier and associates, cleans, and integrates multi-source heterogeneous data according to the physical BOM tree to form an integrated panoramic view of "design-manufacturing-testing" based on the physical object.
[0018] S502: For any selected physical object, the system, based on the constructed equipment commissioning physical object relationship diagram, traces down to the bottom in a complete manner and displays intuitively all its sub-level supporting relationships, basic attributes, production process records, commissioning report sequences and related task execution history, forming a full-dimensional electronic resume of the physical object; S503: When a quality anomaly occurs, a reverse propagation simulation analysis can be initiated based on the batch number or part number of the problematic component. Based on the equipment commissioning physical relationship diagram and historical rules, the parent products and components of all materials loaded with the problematic component batch number or part number can be located, and a list of the scope of impact and the propagation path can be output.
[0019] The advantages of this invention are: (1) Existing technologies often suffer from limitations in timeliness, accuracy, and completeness of query results due to data being scattered across various independent business systems when implementing forward tracing. This results in gaps and incompleteness in tracing information. This invention proposes a time-series attribute-based physical relationship diagram for equipment commissioning, breaking through the limitations of traditional tree structures in expressing complex cross-level and cross-batch relationships. Through path search, the complete assembly and testing process from components to the finished machine can be traced. Furthermore, the system automatically matches and associates metadata with the structured product physical BOM, process route nodes, and commissioning tasks in the data center through the commissioning report association and retrieval module. It also uses the reverse tracing module to quickly locate the source of the problem and its impact range, thereby providing quantitative support for risk assessment and disposal decisions. This solution achieves full-process, high-fidelity, and associative digital tracing and in-depth analysis capabilities.
[0020] (2) This invention addresses the problem that data traceability in existing technologies is usually limited to a single stage (such as manufacturing) and lacks effective integration of data from different stages. It proposes a data center-based end-to-end data connectivity system (covering manufacturing, assembly, testing, and quality inspection). Through data warehouse modeling, it breaks down data barriers between the system and business levels, achieving unified integration and reconstruction of data from multiple stages, including manufacturing, assembly, testing, and quality inspection. Based on this, a two-way intelligent traceability capability is constructed, centered on the physical BOM and combined with a correlation graph. When quality problems occur, the system can fully reconstruct the product's entire lifecycle data and processes based on forward tracing; simultaneously, relying on a reverse impact propagation algorithm, it can quickly and accurately locate all potentially affected similar products. This significantly reduces the time for problem location and impact analysis from several days traditionally required manually to minutes, significantly improving quality response and handling efficiency, and promoting a paradigm shift in quality management from post-event handling to pre-event early warning and precise in-event control. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of the functional module structure of the data center-based commissioning and traceability management system disclosed in the embodiments of the present invention; Figure 2 This is a flowchart of a data center-based commissioning and traceability management system method disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the physical BOM tree corresponding to multi-source heterogeneous data in the data center-based commissioning and traceability management system disclosed in the embodiments of the present invention. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] like Figure 1 and Figure 2 As shown, this embodiment of the invention provides a data center-based commissioning and traceability management system, including: The physical forward traceability module 100, based on a graph structure, constructs a dynamic relationship graph that integrates temporal attributes. It abstracts various entities of electronic equipment (such as product serial numbers, material batch numbers, and commissioning / testing tasks) and their business relationships into nodes and edges. Specifically, it uses the design bill of materials (BOM) and physical component relationships as a foundation, building a dynamic graph data model to overcome the limitations of traditional tree structures in expressing complex cross-level and cross-batch relationships. This model models product structural units, material batches, test nodes, operators, and process versions as nodes, and abstracts the business relationships between entities, such as assembly, testing, and inspection, into directed edges with timestamps and relationship weights, forming a physical relationship graph of equipment commissioning / testing with temporal attributes. A multi-hop association query algorithm is employed for efficient path retrieval, enabling rapid tracing of the complete assembly and testing process from components to the finished product. It supports various dynamic visualization formats such as topology views and timeline views, intuitively displaying complex physical relationships and process sequences. It overcomes the limitations of traditional tree structures in expressing complex, cross-batch relationships, enabling rapid, complete, and visual traceability of massive, high-dimensional commissioning data throughout the entire lifecycle of electronic equipment. The electronic equipment mentioned in this invention includes radar, electronic countermeasures equipment, navigation equipment, and information support equipment. The method proposed in this invention is applicable to all of these equipment domains. This embodiment uses radar equipment as a specific application scenario to provide a detailed description of the proposed method.
[0024] Specifically, the physical forward traceability module 100 based on graph structure association constructs a time-series dynamic graph data model by automatically parsing product design bill of materials, manufacturing bill of materials, and production process data. Each physical unit with a unique identifier (such as an electronic equipment unit with a specific serial number, a transmitter / receiver module, a printed circuit board, and components) is defined as a graph vertex and assigned a physical serial number, material code, and status attribute. At the same time, each recorded assembly operation or test node activity is modeled as a directed edge and assigned attributes including operation type (such as assembly, test), timestamp, operation result, and operator / equipment ID.
[0025] To achieve efficient querying, the system constructs a composite index based on vertex and edge types, that is, a composite index is established for node types (such as components, modules, and complete machines) and edge types (such as assembled in and tested in). When a user initiates a traceability query, the system executes a directed multi-hop graph traversal algorithm oriented towards assembly relationships. It prioritizes exploring paths accelerated by the index along edges of the "assembled in" type, quickly filtering irrelevant branches to efficiently return the complete assembly link from the bottom-level components to the top-level complete machine, and presents it visually in the form of a topology graph. For example, when a user queries the destination of a certain component, the directed multi-hop graph query algorithm can prioritize performing depth- or breadth-first traversal along edges of the "assembled in" type, and use the index to quickly filter irrelevant paths, efficiently returning the complete assembly link from that component to the final integrated machine, and visually displaying it in the form of a topology graph.
[0026] As a further improvement, the physical forward tracing module 100 based on graph structure association proposes a test process visualization tracing that integrates time-series penetration. Starting from the target physical node, it leverages the indexing advantages of the graph database to perform directional traversal along predefined relational edges such as "the object under test," quickly collecting all associated test tasks, equipment, and debugging nodes to form the original relational subgraph. This process is no longer a complex multi-table JOIN operation, overcoming the bottleneck of exponential performance degradation in traditional relational databases during multi-level relational retrieval.
[0027] When tracing the testing history of a specific electronic equipment unit, the time-constrained subgraph query and organization algorithm first extracts all test tasks, equipment, and commissioning nodes related to the electronic equipment unit from the original subgraph. It then extracts the timestamp attributes of the corresponding edges of these test tasks, equipment, and commissioning nodes and performs a global sort. Subsequently, the tracing tasks are grouped according to a configurable strategy (e.g., by day, by test phase), outputting structured timeline data objects. This achieves precise time-series-driven playback, ensuring the absolute accuracy of the test history sequence and providing a directly usable data structure for generating an intuitive timeline view on the front end. Finally, the results are presented in a timeline view, clearly showing when the unit underwent what tests, which equipment was used, and the results, enabling dynamic playback of the test history.
[0028] The data center-based commissioning report association and retrieval module 200 automatically extracts key entities such as product serial numbers, test items, times, and commissioning equipment from various reports using an unstructured document parsing engine, and transforms them into standardized metadata. Relying on a pre-built business rule engine, the system automatically matches and associates the metadata with the structured product physical BOM, process nodes, and commissioning tasks within the data center, constructing a multi-dimensional index system of "report-physical-process stage." Commissioning process files, result files, and reports are centrally stored on a file server, while the standardized metadata and associated indexes are stored in a high-performance database, supporting users to perform fast and accurate retrieval and full-process traceability using multi-dimensional conditions such as product serial number, time range, and test type.
[0029] Specifically, the data center-based commissioning report association and retrieval module 200, through the configuration of an extensible, business logic-based business rule base, explicitly defines the semantic mapping strategy between report metadata and product BOM nodes and process tasks. That is, it explicitly defines how to map report metadata (such as product serial numbers, test item codes, and timestamps) to structured business objects within the data center, achieving automatic and accurate association between commissioning reports and structured business objects. Through a rule matching engine, it automatically performs logical matching and reasoning with pre-defined rules to establish a bidirectional association between the report, physical objects, and node tasks, forming an entity relationship network that supports efficient retrieval and traceability. For example, if the product serial number in the report metadata completely matches the serial number field of a node in the product's physical BOM, and the report's test time falls between the planned start time and actual completion time of a commissioning task in the process route for that node, then a bidirectional association is automatically established between the report, the physical node, and the commissioning task node. When the business rule matching engine is running, it can automatically match and reason with the standardized report metadata and these rules to achieve accurate and automatic mounting between the commissioning report, physical products, and node tasks, forming a machine-readable association network.
[0030] The reverse tracing module 300 based on impact propagation analysis, according to the constructed equipment commissioning physical relationship diagram, combined with material loading relationships, test dependencies, and historical fault association rules, reverse analyzes and calculates a list of all potentially affected upper-level assemblies and final products, and marks the propagation path of the impact (e.g., capacitor → power board → control unit → complete machine) and confidence level, rather than simply listing parent items level by level. It supports user-defined tracing depth (e.g., tracing only to the module level or up to the complete machine) and association rules (e.g., tracing only loading relationships within a specific time range) to quickly locate the impact range of the problem source and provide quantitative basis for risk assessment and handling decisions.
[0031] Specifically, the reverse tracing module 300 employs a rule-based and statistical reverse influence propagation algorithm. When a user selects a faulty component to initiate reverse tracing, the system starts with the faulty component and performs a reverse graph traversal and influence propagation simulation in the equipment commissioning physical relationship diagram. The system defines differentiated propagation rules and weights for different types of edges: for structural relationship edges such as "assembled in," a weight of 1.0 is assigned, indicating that the influence will inevitably propagate along this path; for statistically related edges such as "test dependent," the conditional probability is calculated based on historical data as the weight. The details are as follows: 1) Forced propagation edge (e.g., "assembled in"): with a weight of 1.0, the influence will inevitably propagate back to the parent node along this path.
[0032] 2) Probabilistic propagation edges (e.g., "test dependencies"): Weights are derived from historical data statistics (e.g., if test A fails, test B has an 80% probability of also failing, then the propagation weight of edge A→B is...). 。 The confidence level is calculated as follows: The algorithm starts from the source node and propagates layer by layer along the reverse edges. Upon reaching each parent node, it dynamically calculates the affected confidence level based on the current node's confidence level, the weight of the propagation edge, and the path decay factor. The affected confidence level is calculated as follows: Starting from the source node of the fault, traverse along the reverse edges. When the fault affects a secondary node... (Confidence level) ) through a weight of The edge propagates to its parent node. hour, The affected confidence level obtained through this path is calculated using the following formula:
[0033] in, : Current node The confidence level affected.
[0034] From node To the parent node The propagation edge weights are derived from historical data statistics.
[0035] Path decay factor ( ).
[0036] Total number of hops in the current propagation path (from the source node to the destination node). (Number of sides).
[0037] A single upstream node may be affected through multiple downstream paths, and its overall confidence level of being affected is taken as the maximum value calculated across all propagation paths. To ensure that key risks are not underestimated.
[0038] At the end of the tracing process, the system outputs structured analysis results and an intelligent analysis report, including an impact list sorted by confidence level, a visualized propagation map, and high-risk handling recommendations filtered based on confidence thresholds. This enables a quantitative assessment and precise location of the fault's impact scope. Details are as follows: 1) Impact List: All affected upper-layer products / modules, sorted from highest to lowest confidence level.
[0039] 2) Propagation map: Visualizes the main paths of influence diffusion.
[0040] 3) Key recommendations: Automatically identify "high-risk items" with a confidence level higher than a preset threshold (e.g., 0.9) and recommend prioritizing their detection or handling.
[0041] The reverse tracing module 300, based on impact propagation analysis, utilizes graph computing and probabilistic reasoning methods to achieve quantitative risk assessment and precise location from the fault point to the impact domain. This module uses a physical relationship diagram of equipment commissioning and testing as its foundation. Through a reverse traversal algorithm, combined with historical weights and propagation rules of relationships such as material loading, test dependencies, and process constraints, it calculates the probability of impact and propagation path of the problematic component on all superior assemblies and the final product. It supports custom configuration of tracing depth (e.g., sub-components, other parts) and time range (e.g., the past month), and outputs an impact list with confidence assessment and a visualized propagation map to help users assess the scope of fault impact and create branch commissioning sub-tasks.
[0042] By constructing a time-series subgraph centered on test tasks, the visualization and traceability of the debugging process can be achieved. Specifically, each debugging task (such as sensitivity testing) is modeled as an independent node, and directed edges of types such as the object under test, the equipment used, and the standard used are used to associate it with the corresponding physical electronic equipment, test equipment, and debugging reference document nodes. Each edge is assigned a precise timestamp attribute.
[0043] When tracing the testing history of a specific electronic equipment unit, the system executes a time-constrained subgraph query algorithm. First, it locates the target unit node. Then, it retrieves all associated test task nodes along the edge of the "object under test," and strictly sorts the task nodes according to their time sequence and groupes them by process stage based on the edge timestamps. Finally, the system dynamically presents the results in a timeline view, clearly displaying the time nodes, test items, equipment used, and result status of the unit in each test, completely reproducing its testing history.
[0044] Continue reading Figure 2This invention also provides a method for constructing a data center-based commissioning and traceability management system, comprising the following steps: S501: Full-element data acquisition and unified modeling based on physical BOM. The data center connects with business systems across all stages of the electronic equipment product development lifecycle—design, process, production, and testing—to achieve full acquisition of structured data and unstructured documents and records. The system uses the product's physical serial number as a unique identifier and automatically associates, cleans, and integrates multi-source heterogeneous data according to the physical BOM tree, constructing an integrated panoramic data view of "design-manufacturing-testing" with the physical product as the main thread. The specific principle and process of associating, cleaning, and integrating multi-source heterogeneous data according to the physical BOM tree are as follows: The core is to build an association framework with the product's physical serial number as the root key, an extended BOM tree as the skeleton, and event sequence and business rules as the connection logic.
[0045] By utilizing data centers, standardized access and preliminary analysis of data enable the full collection of structured data and unstructured documents and records across all stages of the electronic equipment product development lifecycle, including design, process, production, and testing.
[0046] Simultaneously, the system forcibly extracts or infers the physical product serial number from each record / file. For explicitly included information, it extracts it directly; for missing information, it uses contextual association rules to complete it. For example, by matching the timestamp and equipment number of the test record with the production execution plan, it reversely completes the serial number of the products processed by that equipment during that period.
[0047] The dynamic product physical BOM tree not only includes the material composition of the design BOM (multiple levels such as whole machine → unit → module → board, etc.), but also extends to the physical serial number and production path information of each drawing number. The rule engine uses the extracted "serial number" as the primary key to automatically mount the processed data package (containing data content, metadata, and file index) to the corresponding node in the physical BOM tree. After clarifying the association context, data cleaning is performed using data quality services, such as unit conversion, value range verification, and comparative analysis with design parameters. Using classification and hierarchical modeling, a complete data view of a specific electronic equipment product is reconstructed through the serial number and BOM node ID. All data, regardless of its original format, will ultimately find its definite location and establish association relationships within this framework, such as... Figure 3 As shown.
[0048] S502: Forward Holographic Traceability Based on Relationship Graph. For any selected physical object (such as a complete electronic equipment unit or module), the system automatically performs multi-hop association queries based on the constructed physical object relationship graph. Through this step, it is possible to trace down completely and intuitively display all its sub-level supporting relationships, basic attributes, production process records, commissioning report sequences, and related task execution history, forming a full-dimensional electronic resume for the physical object.
[0049] As a further improvement, to adapt to the query and analysis habits of different roles, the system supports organizing and presenting data in four modes: design BOM view, physical BOM view, professional dimension view, and file type view. Users can perform global drill-down and visualization linkage on BOM nodes, related files, and structured parameters in any view, significantly improving the efficiency of exploring complex data relationships.
[0050] S503: Precise Reverse Positioning Based on Impact Propagation. When a quality anomaly occurs, a reverse impact propagation analysis algorithm can be activated based on the batch number or part number of the problematic component. This algorithm, based on physical relationship maps and historical rules, quickly locates all upstream products and components containing materials of that batch / part number, and outputs a list of affected areas and propagation paths, enabling precise targeting of the problem's scope and comprehensive, inductive management.
[0051] As a further improvement, the system automatically generates a structured output based on user-defined traceability conditions. This output includes a list of affected parent physical items, a visualized propagation path, and a quantified risk level, enabling precise definition and proactive control of the problem scope. The output list comprehensively covers key traceability attributes, such as the top-level physical item serial number, name, drawing number, order number, batch number and name of associated components, and complete matching relationships. This provides users with a clear and complete overview of the impact scope, significantly assisting subsequent decision-making and handling.
[0052] Through the above technical solutions, this invention addresses the problem of broken traceability chains and distorted analysis and comparison caused by the heterogeneous data formats and lack of effective correlation among multi-source heterogeneous data in existing testing equipment. The solution utilizes the following technical means: 1. For various types of data generated during the commissioning and testing of electronic equipment, such as processing inspection, production / assembly, debugging, and rework data, the system provides a configurable data acquisition adapter. This adapter supports automatic parsing and access to structured, semi-structured, and unstructured data in various formats, including .dat, .xml, database records, videos, and images. Simultaneously with data access, key information cleaning and completion rules are executed to ensure that each data entry possesses core attributes such as a correlated product serial number, material code, timestamp, and process identifier, laying a standardized data foundation for subsequent data correlation and traceability. 2. Based on a unified data model built within the data center, using the material code as a globally unique index, the cleaned data is automatically correlated to the corresponding physical product nodes, process steps, and commissioning task instances, forming a structured data chain with "physical object—process—task" as the main thread. Through a rules engine and automatic matching algorithm, the system intelligently binds test results, report files, and underlying process data, ensuring complete traceability of data at both the semantic and business levels. Third, based on the associated data, the system dynamically generates a digital relationship graph of electronic equipment products with time-series attributes, intuitively presenting complex relationships across processes, batches, and levels. Based on this graph, it possesses traceability capabilities for both forward end-to-end traceability and reverse impact analysis.
[0053] To address the challenges of inconsistent standards for recording key parameters and missing data fields in existing technologies, which hinder effective correlation analysis, this invention systematically solves the problem by establishing a unified data cleaning and information completion mechanism: First, the system incorporates a data quality detection engine that automatically identifies missing key fields in each record, such as product identifier, node name, operator, equipment number, and timestamp. Based on a predefined business rule base, it automatically infers and completes missing fields. For example, using timestamp alignment technology, it matches test records with product commissioning plans to complete product model and serial number information; and automatically associates operators and commissioning nodes based on equipment-process-personnel relationships and task execution time periods. Second, a unified time-series event model is established in the data center, aligning and integrating multi-dimensional data such as test records, task execution logs, equipment status, and personnel operation records on a timeline. Using an event correlation model, it automatically establishes a complete link between "product-node-personnel-equipment-parameter-result-time," forming structured and traceable parameter records. Third, standardize design parameters and unify metadata modeling. Establish standardized naming and measurement specifications for various test parameters (such as performance indicators and environmental data) to ensure semantic consistency of data across different devices and batches. Through a metadata-driven data model, define the source, unit, valid range, and associated objects of each parameter, making it identifiable, associatable, and analyzable within the system. Fourth, utilize data quality monitoring in the data center to support continuous monitoring of data integrity and accuracy, and generate data quality reports. Manual assistance is provided to correct incomplete data records for certain parameters. This ensures the usability and reliability of parameter records in subsequent traceability, statistics, and comparisons.
[0054] To address the issues of delayed decision-making and incomplete root cause identification caused by lagging process data updates, the previous reliance on manual data retrieval has been replaced by more automated system data collection. For data requiring manual input, operators simply need to save the resulting files (such as Excel files or text reports) to a pre-defined standardized collection directory. The system automatically detects new files and triggers the parsing process through folder monitoring services, eliminating the need for manual data transfer. For systematic data from MES, testing equipment, etc., automatic retrieval or real-time reception is achieved through adapter interfaces, recording complete data context (such as generation time, equipment ID, and task ID). The system then automatically extracts key information based on pre-configured parsing templates and a rule engine, such as identifying product models, serial numbers, test nodes, and timestamps from filenames, file headers, or specific fields. The extracted data is automatically matched and mounted against the physical tree structure of electronic equipment products in the data center using key identifiers such as material codes, serial numbers, and process codes. This establishes a dynamic relationship between "data-physical objects-process nodes-test tasks," ensuring that each piece of data is locatable and semantically understandable within the system. Such a large amount of data processing work can be carried out in parallel in the background, ensuring timely data updates.
[0055] In summary, the data center-based commissioning and traceability management method proposed in this invention relies on the data center to achieve unified access, standardized integration, and centralized management of multi-source and heterogeneous data throughout the entire product development process. It covers various types of data in product design, process, manufacturing, materials, quality, and testing. With complete and reliable data support, the system can achieve traceability and reusability of data throughout the entire product lifecycle, and provide feedback for continuous optimization of the product realization process.
[0056] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A data center-based commissioning and traceability management system, characterized in that, Includes the following modules: The forward traceability module is used to model entities such as product structural units, material batches, test nodes, operators, and process versions as nodes, and to abstract the business relationships between entities, such as assembly, testing, and inspection, as directed edges with timestamps and relationship weights, forming a physical relationship graph of equipment commissioning with time sequence attributes; based on the physical relationship graph of equipment commissioning, the complete assembly and testing process from components to the whole machine can be traced through path search. The commissioning report association and retrieval module is used to extract key entities such as product serial number, test item, time, and commissioning equipment from various commissioning reports, and convert them into standardized metadata. This standardized metadata is then automatically matched and associated with the structured product physical bill of materials, process route nodes, and commissioning tasks in the data center, thereby building a multi-dimensional index relationship between report, physical object, and process stage. Users can search and trace products using key fields such as product serial number, time range, and test type. The reverse tracing module is used to reverse analyze and calculate all potentially affected upper-level assemblies and final product lists based on the constructed equipment commissioning physical relationship diagram, combined with material loading relationships, test dependencies, and historical fault association rules. It also marks the propagation path and confidence level of the impact, and prioritizes the detection or disposal of upper-level assemblies with confidence levels higher than a preset threshold.
2. The data center-based commissioning and traceability management system according to claim 1, characterized in that, The forward tracing module is also used for: By analyzing various data from the design bill of materials, manufacturing bill of materials, and production execution process of electronic equipment products, a dynamic graph data model is constructed. This model models each uniquely identified physical unit as a graph node, assigning it attributes such as physical serial number, material code, and status. Simultaneously, each recorded assembly operation or test node activity is modeled as a directed edge, assigning attributes including operation type, timestamp, operation result, and the ID of the operator or equipment. A composite index is established for node and edge types. When a user queries the destination of a component corresponding to a specific node type, a depth- or breadth-order traversal is performed along the edges corresponding to that edge type. Irrelevant paths are filtered using the index, returning the complete assembly chain from the component to the final integrated machine, and visually displayed in the form of a topology graph.
3. The data center-based commissioning and traceability management system according to claim 2, characterized in that, The physical unit with independent identification includes an electronic equipment unit with a serial number, a transmitting or receiving module, a printed circuit board, and components; the operation type includes assembly and testing.
4. The data center-based commissioning and traceability management system according to claim 3, characterized in that, The node types include components, modules, and complete machines, and the edge types include assembly and testing.
5. The data center-based commissioning and traceability management system according to claim 1, characterized in that, The forward tracing module is also used for: Starting from the target physical node, a directed traversal is performed along predefined relational edges using a graph database to collect all test tasks, equipment, and debugging nodes associated with the target physical node, forming an original relational subgraph. Here, predefined relational edges refer to relational edges with the target physical node as the object under test. When tracing the testing history of a complete electronic equipment, all test tasks, equipment, and debugging nodes related to the complete electronic equipment are first extracted from the original relational subgraph. The timestamp attributes of the corresponding edges of the test tasks, equipment, and debugging nodes are then extracted and globally sorted as tracing tasks. Subsequently, the tracing tasks are grouped by day or by test stage. Finally, the results are presented through a timeline view, showing when the complete equipment underwent what kind of test, which equipment was used, and what the result was, realizing dynamic playback of the test history.
6. The data center-based commissioning and traceability management system according to claim 1, characterized in that, The commissioning report association and retrieval module is also used for: Configure the association rule base to define how to map report metadata to structured business objects in the data center. The mapping process is as follows: if the product serial number in the report metadata matches the serial number field of a physical node in the product's physical bill of materials exactly, and the report test time is between the planned start time and the actual completion time of a commissioning task in the process route for that node, then establish a bidirectional association between the report and the physical node and the commissioning task node.
7. The data center-based commissioning and traceability management system according to claim 6, characterized in that, The report metadata includes the product serial number, test item code, and timestamp.
8. The data center-based commissioning and traceability management system according to claim 1, characterized in that, The reverse tracing module is also used for: When a user selects a faulty component to initiate reverse tracing, a reverse propagation simulation analysis is performed starting from the faulty component in the equipment commissioning physical relationship diagram. The reverse propagation simulation analysis starts from the source node and traverses along the reverse edge. The confidence level of each parent node reached is calculated as follows: ,in, For the current node The impact confidence level parent node confidence level For the node To the parent node The propagation edge weights, The path decay factor, For from the source node to The number of edges, the comprehensive impact confidence of each parent node is the maximum value of the impact confidence calculated in all propagation paths; all propagation paths are recorded synchronously, and an intelligent analysis report is output at the end of the tracing. The intelligent analysis report includes an impact list of all affected upper-level products or modules sorted from high to low confidence, a propagation map of the impact diffusion path, and key recommendations for priority detection or disposal of upper-level assemblies with confidence levels higher than the preset threshold.
9. The data center-based commissioning and traceability management system according to claim 8, characterized in that, The propagation edge weights are obtained as follows: In the physical relationship graph of equipment commissioning, the weight of the edge is set to 1 for the influence that will inevitably propagate back to the parent node along the path. For the edge with test dependency, the weight is set according to the test dependency relationship. For example, if test A fails, test B has an x% probability of also failing, then the weight of the propagation edge between the two is 0.0x.
10. The method for a data center-based commissioning and traceability management system according to any one of claims 1-9, characterized in that, include: S501: The data center connects to the business systems of all stages in the entire lifecycle of electronic equipment product development, including design, process, production, and commissioning. It achieves full collection of structured data and unstructured documents and records. The system uses the product's physical serial number as a unique identifier and associates, cleans, and integrates multi-source heterogeneous data according to the physical BOM tree to form an integrated panoramic view of "design-manufacturing-testing" with the physical object as the main line. S502: For any selected physical object, the system, based on the constructed equipment commissioning physical object relationship diagram, traces down to the bottom in a complete manner and displays intuitively all its sub-level supporting relationships, basic attributes, production process records, commissioning report sequences and related task execution history, forming a full-dimensional electronic resume of the physical object; S503: When a quality anomaly occurs, a reverse propagation simulation analysis can be initiated based on the batch number or part number of the problematic component. Based on the equipment commissioning physical relationship diagram and historical rules, the parent products and components of all materials loaded with the problematic component batch number or part number can be located, and a list of the scope of impact and the propagation path can be output.