Digital resource sharing method and system based on pedigree mapping relation

Abstract

The invention relates to a digital resource sharing method and system based on a pedigree mapping relation, and belongs to the technical field of enterprise digital resource management, and the resource sharing method comprises the steps: obtaining multi-source heterogeneous data from an internal distributed system of an enterprise through a preset interface protocol; performing standardization processing on the multi-source heterogeneous data to generate a structured resource pool; a graph dictionary module is constructed, the resources in the structured resource pool are subjected to association topology analysis, and a resource relation topological graph is output; establishing a pedigree relationship between resources and business scenes based on the three-dimensional mapping model, respectively generating and fusing a business process demand map, a production stage demand map and a product capability matching map, and constructing a visual digital resource capability platform; and outputting the target resource identifier and the associated path in response to a resource calling instruction input by a user. Integration and sharing of internal and external multi-source heterogeneous data of an enterprise can be realized, the problem of data islands is solved, and meanwhile, the intelligent degree of a resource management system is improved.

Description

Technical Field

[0001] This application relates to the technical field of enterprise digital resource management, and in particular, to a digital resource sharing method and system based on pedigree mapping relationships. Background Art

[0002] In recent years, with the rapid development of information technology, digital transformation has become the core strategy for global enterprises to enhance competitiveness, optimize resource allocation, and achieve sustainable development. Especially in industries such as manufacturing, services, and finance, enterprises have gradually digitized traditional business processes and promoted the intelligent management of various businesses with the help of advanced technologies such as the Internet, Internet of Things, big data, and artificial intelligence. However, in the process of promoting digital transformation, enterprises are also facing an increasingly prominent problem - the integration and sharing of multi-source heterogeneous data.

[0003] In practical applications, enterprises often use multiple information systems to handle various business requirements. For example, personnel management in enterprises usually relies on the HR system, business project management relies on the project management system, and customer relationship management relies on the CRM system, etc. In addition, with the expansion of enterprise business, the addition of external systems and partners makes the data sources more diverse. The various data sources of enterprises not only include traditional structured data (such as tabular data in databases), but also cover semi-structured and unstructured data (such as documents, logs, reports, sensor data, etc.).

[0004] Currently, when existing enterprise resource management systems process these different types of semi-structured and unstructured data across systems and domains, due to inconsistent formats and different storage methods, there is often a lack of effective association between data, resulting in the emergence of information island problems. Moreover, common resource recommendation systems mostly rely on static rules or keyword-based matching methods, lacking the ability to reason about the deep associations and semantics between data, resulting in poor resource matching efficiency and cross-database resource integration ability, thereby restricting the utilization value of resources and making it difficult to meet the rapidly developing business needs of enterprises. Summary of the Invention

[0005] In order to achieve the integration and sharing of multi-source heterogeneous data inside and outside the enterprise, solve the data island problem, and at the same time improve the intelligence level of the resource management system, this application provides a digital resource sharing method and system based on pedigree mapping relationships.

[0006] In the first aspect, this application provides a digital resource sharing method based on pedigree mapping relationships, adopting the following technical solution: A digital resource sharing method based on pedigree mapping relationships, the resource sharing method includes: Obtain multi-source heterogeneous data from the enterprise's distributed system through a preset interface protocol; Standardize the multi-source heterogeneous data based on a preset resource classification standard to generate a structured resource pool; Construct a graph dictionary module to perform associated topological analysis on the resources in the structured resource pool and output a resource relationship topology graph; Based on the structured resource pool and the resource relationship topology graph, establish a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model, and generate a business process requirement graph, a production stage requirement graph, and a product ability matching graph respectively; Fuse the business process requirement graph, the production stage requirement graph, and the product ability matching graph to construct a visual digital resource capability platform; Respond to a resource call instruction input by a user, and output a target resource identifier and an associated path based on the visual digital resource capability platform.

[0007] By adopting the above technical solutions, a unified pedigree mapping relationship is constructed to eliminate data islands. A structured resource pool is constructed by means of ontology semantic modeling, and a resource relationship topology is generated by using a dynamic graph computing engine. By innovatively introducing a three-dimensional tensor pedigree, business requirements, resource supply, and product capabilities are modeled under a unified mathematical framework. Through multi-graph fusion and visual analysis, a global perspective of enterprise digital resources is realized. Finally, an accurate resource path is output according to the user's resource call requirements. This application realizes the integrated management, standardized application, and value empowerment of digital resources, constructs an enterprise-level digital resource capability platform covering multiple levels and scenarios, and can provide strong support for the digital transformation of enterprises.

[0008] In a second aspect, this application provides a digital resource sharing system based on a pedigree mapping relationship, adopting the following technical solutions: A digital resource sharing system based on a pedigree mapping relationship, the resource sharing system includes: A heterogeneous data acquisition module, configured to acquire multi-source heterogeneous data from an enterprise internal distributed system through a preset interface protocol; A data processing module, configured to standardize the multi-source heterogeneous data based on a preset resource classification standard to generate a structured resource pool; An associated topological analysis module, configured to construct a graph dictionary module to perform associated topological analysis on the resources in the structured resource pool and output a resource relationship topology graph; A multi-graph generation module, configured to establish a pedigree relationship between resources and business scenarios based on the structured resource pool and the resource relationship topology graph, and generate a business process requirement graph, a production stage requirement graph, and a product ability matching graph respectively; A platform construction module, which is used to integrate the business process requirement map, the production stage requirement map, and the product capability matching map to construct a visual digital resource capability platform; A target resource identifier output module, which is used to respond to a resource call instruction input by a user and output a target resource identifier and an association path based on the visual digital resource capability platform.

[0009] Thirdly, the present application provides a computer device, adopting the following technical solution: A computer device includes a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps of the method described in the first aspect.

[0010] Fourthly, the present application provides a computer-readable storage medium, adopting the following technical solution: A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to perform any one of the methods in the first aspect.

[0011] In summary, the present application includes at least one of the following beneficial technical effects: dynamically integrating dispersed multi-source heterogeneous data within an enterprise through a preset interface protocol to eliminate information silos; converting raw data into a structured resource pool based on a unified classification standard to achieve standardization and normalization of resource descriptions; parsing the topological associations between resources with the aid of a map dictionary module to generate a visual relationship network; accurately constructing three core maps of business process requirements, production stage requirements, and product capability matching based on a three-dimensional mapping model to form a genealogical mapping relationship between resources and scenarios; integrating multi-dimensional maps to construct a visual platform to achieve global perspective and dynamic management of resource status; and finally, through an intelligent response mechanism, accurately positioning a target resource and its association path according to a user instruction.

[0012] The present application significantly improves the resource retrieval efficiency, strengthens the cross-scenario matching accuracy, constructs a digital resource center with agility, accuracy, and security for an enterprise, and thus comprehensively enables business collaborative innovation and the implementation of the digital transformation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 FIG. is a first process schematic diagram of a digital resource sharing method based on a genealogical mapping relationship in one embodiment of the present application.

[0014] Figure 2 FIG. is a structural schematic diagram of a three-dimensional mapping model in one embodiment of the present application.

[0015] Figure 3 FIG. is a second process schematic diagram of a digital resource sharing method based on a genealogical mapping relationship in one embodiment of the present application.

[0016] Figure 4 It is the third process schematic diagram of the digital resource sharing method based on the pedigree mapping relationship in one embodiment of the present application.

[0017] Figure 5 It is the fourth process schematic diagram of the digital resource sharing method based on the pedigree mapping relationship in one embodiment of the present application.

[0018] Figure 6 It is the fifth process schematic diagram of the digital resource sharing method based on the pedigree mapping relationship in one embodiment of the present application.

[0019] Figure 7 It is the sixth process schematic diagram of the digital resource sharing method based on the pedigree mapping relationship in one embodiment of the present application.

[0020] Figure 8 It is the seventh process schematic diagram of the digital resource sharing method based on the pedigree mapping relationship in one embodiment of the present application. Detailed implementation manners

[0021] In order to make the purpose, technical solutions and advantages of the present application clearer and more understandable, the following Figure 1-8 in combination with the appended drawings and embodiments, the present application will be further described in detail. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0022] An embodiment of the present application discloses a digital resource sharing method based on the pedigree mapping relationship.

[0023] Referring to Figure 1 , a digital resource sharing method based on the pedigree mapping relationship, the resource sharing method includes: Step S101, obtain multi-source heterogeneous data from the enterprise internal distributed system through a preset interface protocol; Among them, the multi-source heterogeneous data includes but is not limited to personnel management data, case selection data, and business scenario project data; the main sources include the company's digital personnel management system, digital case selection system, and "Hundred Scenarios" system. The "Hundred Scenarios" system is mainly used to centrally display the key business scenarios and successful cases implemented during the company's digital transformation.

[0024] Specifically, preset interface protocols (such as RESTful API, JDBC direct connection protocol) are essentially a set of predefined communication rules and data structure conversion standards, which can automatically identify the data formats of source systems (such as ERP, CRM, document management systems) through a protocol parsing engine. For example, when accessing the "SAP HR system", the protocol adaptation layer will convert the IDOC format into a standardized JSON-LD (Linked Data) structure to ensure that the organizational structure relationships in personnel management data can be parsed; for unstructured data (such as construction log texts), key entities (such as project numbers, equipment models) will be automatically extracted through an NLP entity recognition engine and metadata tags will be injected. The aggregation of multi-source heterogeneous data depends on a dynamic routing mechanism, that is, the optimal transmission path (such as Kafka message queue for real-time data transmission, FTP for batch document transmission) is automatically selected according to the data source type (structured SQL tables, semi-structured MongoDB documents, unstructured PDFs). This process can achieve non-intrusive data integration between internal and external enterprise systems and avoid destructive transformation of the original system architecture.

[0025] It can be understood that building a unified data access plane can convert heterogeneous information such as personnel qualification data scattered in the human resources system, innovative scoring data in the case selection system, and business scenario metadata in the "hundred major scenarios" system into spatio-temporally aligned atomic data units, eliminating data semantic ambiguity caused by system isolation at the source. For example, after the data of a certain tunnel monitoring device is protocol-converted, fields such as its timestamp, location coordinates, and sensor type can be consistently parsed across systems, providing standardized input for subsequent resource pool construction; Step S102, based on a preset resource classification standard, perform standardization processing on multi-source heterogeneous data to generate a structured resource pool; Specifically, the predefined resource classification standard is actually a domain ontology model. For example, the OWL (Web Ontology Language) class hierarchy relationship with the business attribute dimension defined as "technology - case - tool" and the three-level concept system of "group → secondary unit → project" of SKOS (Simple Knowledge Organization System) is adopted for the hierarchical dimension.

[0026] Exemplarily, when multi-source heterogeneous data is input, the ontology reasoning engine performs semantic alignment and instantiation: automatically matching the "innovation case application form" field in the data unit to the case class resources in the benchmark library, and mapping the "bridge health monitoring case" label of this resource to the "highway domain sub-library" (identifying text topic similarity based on the LabelRank algorithm). Knowledge base construction depends on concept graph extraction. For example, "tunnel support technology" is extracted from technical specification documents as a knowledge node, and an "technology - material" association edge is established with the concrete mix ratio table.

[0027] In addition, the structured resource pool contains independent benchmark libraries, knowledge bases, product libraries, and expert libraries. The isolation of the four resource libraries (benchmark library, knowledge base, product library, expert library) is achieved through resource partition namespaces. For example, the expert library uses the domain type as the shard key (such as hash value sharding) to ensure that water transportation expert information is not miswritten into the construction domain library.

[0028] Step S103: Construct a graph dictionary module to perform associated topological analysis on the resources in the structured resource pool and output a resource relationship topological graph. Among them, the graph dictionary module is essentially a dynamic graph computing system, and its core includes: (1) Topological relationship miner: Based on the attribute tags of the resources in the resource pool (such as the "applicable formation" field of "shield construction method" in the knowledge base), calculate the semantic distance between resources (using the TransE embedding model to minimize the spatial vector difference of the "method - equipment" triple); (2) Relationship weight quantizer: Use the Structural Entropy algorithm to analyze the influence of resource nodes in the business context. For example, a high-weight connection edge (weight value = log(joint call times)) is generated between the frequently co-called "monitoring and measurement software" and "tunnel convergence algorithm". The generation of the topological graph depends on the dynamic graph convolutional network. When a new "diaphragm wall leakage treatment report" is added, the resource node is automatically embedded into the existing graph (edge generation rule: if the text similarity with the "foundation pit support specification" > 0.7, then an edge is established), and finally a JSON graph structure with an adjacency matrix is output.

[0029] It can be understood that the resource relationship topological graph transforms discrete resources into a globally connected semantic network. Examples show that when a user retrieves "bridge crack repair", the system takes the "crack repair technology" node as the center in the topological graph and automatically expands associated resources along the relationship edges (weight sorting: repair material library → detection equipment manual → expert directory), upgrading the resource discovery process from keyword matching to semantic reasoning. Experimental data show that this mechanism increases the associated recall rate of cross-library resources by 62%, and due to the dynamic embedding mechanism, the newly added "AI crack recognition model" can automatically form an association chain with existing cases and tools.

[0030] Step S104: Based on the structured resource pool and the resource relationship topological graph, establish a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model, and generate a business process requirement graph, a production stage requirement graph, and a product capability matching graph respectively. Among them, the three-dimensional mapping model is a high-order tensor, divided into the demand side (X-axis and Y-axis) and the supply side (Z-axis). The X-axis and Y-axis accurately reflect the resource requirements of each professional project within the company, and the Z-axis shows the existing and potential digital product resources of the enterprise.

[0031] Specifically, the business process requirements graph (X dimension): Model the spatio-temporal relationships of the business process chain. For example, break down the construction stage into a [process node × time slot] matrix, and the dependency relationships between nodes are calculated by the Petri net transfer function; for example, if the output delay of the "pile foundation inspection report" exceeds the threshold, trigger an alarm for the "construction suspension" node.

[0032] Production stage requirements graph (Y dimension): Describe the capacity coverage of resources during the production cycle. Predict resource requirements through a hidden Markov model. For example, during the typhoon season, the probability of calling the "flood control expert library" increases by 85%, and construct a resource-life cycle alignment matrix.

[0033] Product capability matching graph (Z dimension): Quantify the fit between resources and scenarios. Introduce a collaborative filtering algorithm. For example, when the "displacement warning accuracy > 90%" of a certain slope monitoring radar in multiple tunnel projects, increase its scoring factor on the Z axis , where score is the finally obtained scoring factor, α is the weight coefficient of the call times, and β is the weight coefficient of successful cases.

[0034] Based on the above model architecture, the three-dimensional graph fusion relies on tensor chain decomposition. By decomposing high-order relationships into the product of low-rank core tensors, the computational complexity is reduced. Exemplarily, the practical application of the three-dimensional pedigree in the bridge monitoring scenario shows that: the X-dimensional demand chain accurately locates the "post-rainstorm monitoring" sub-process, the Y-dimensional prediction requires calling the "radar monitoring equipment" (weather forecast trigger probability threshold), and the Z-dimensional automatically matches the "millimeter wave radar P12" in the equipment library (score > 97% of the same type). The tensor operation of the three reduces the resource matching path length to an average of 2.1 hops, which can improve the efficiency by 3 times compared with the traditional solution.

[0035] Step S105, fuse the business process requirements graph, the production stage requirements graph, and the product capability matching graph to construct a visual digital resource capability platform; Among them, the essence of platform construction is the integration of a graph visualization analysis engine and a distributed computing framework.

[0036] Specifically, the fusion process adopts multi-graph joint embedding, and the specific steps include: (1) Calculate the optimal alignment matrix between the nodes of the demand-side graph (business process requirements graph and production stage requirements graph) and the resource nodes of the supply-side graph (product capability matching graph) through the Gromov-Wasserstein distance; (2) Render the optimal alignment matrix as a three-dimensional force-directed graph based on the D3.js engine. The node size in the graph is scaled logarithmically according to the resource call frequency, and the edge width is linearly enlarged according to the collaborative filtering score; (3)When new business requirements (such as the need for intelligent construction site monitoring) arise, locate the affected sub-graphs of the graph and locally reconstruct the alignment matrix and rendering results. Specifically, the local graph can be reconstructed by calling TensorFlow (only re-decompose the affected sub-graphs) to avoid the resource consumption of global recalculation.

[0037] Step S106, in response to the resource call instruction input by the user, output the target resource identifier and the associated path based on the visual digital resource capability platform.

[0038] Specifically, the instruction response can be based on a path planning algorithm driven by reinforcement learning. When the user initiates a resource call (such as "obtain coastal soft foundation treatment cases"), the system can output an accurate resource path through a deep reinforcement learning strategy. Exemplarily, in a port reconstruction and expansion project, the user requests an "automated terminal scheduling plan": after the system locates the "automated terminal benchmark library" node in the topology graph, it automatically outputs [port scheduling algorithm SDK] → [device interface protocol] → [list of external experts] along the high-weight associated path.

[0039] In the above embodiments, a unified pedigree mapping relationship is constructed to eliminate data islands. With the help of ontology semantic modeling, a structured resource pool is constructed. Using a dynamic graph computing engine, a resource relationship topology is generated. By innovatively introducing a three-dimensional tensor pedigree, business requirements, resource supply, and product capabilities are modeled under a unified mathematical framework. Through multi-graph fusion and visual analysis, a global perspective of enterprise digital resources is achieved. Finally, an accurate resource path is output according to the user's resource call requirements. This application realizes the integrated management, standardized application, and value empowerment of digital resources, constructs an enterprise-level digital resource capability platform covering multiple levels and scenarios, and can provide strong support for the digital transformation of enterprises.

[0040] Refer to Figure 2 As shown, it is a three-dimensional mapping model constructed in one embodiment of this application. The three-dimensional structure is composed of the X-axis, Y-axis, and Z-axis, each representing a different dimension. The X-axis focuses on the mining of specific business processes. The nodes on the axis are connected into lines, which can comprehensively cover the typical processes and key scenarios of all professional engineering. The Y-axis represents specific production nodes, and the node connections run through the entire life cycle from planning and design to operation and maintenance. The Z-axis represents specific digital products, and the node connections cover all valuable digital products of the company.

[0041] Among them, the demand side is jointly constructed by the X-axis and the Y-axis, which can accurately reflect the resource requirements of each professional project within the company. The business map comprehensively sorts out and presents the specific links and key requirements of project implementation according to the business process shown on the X-axis; the demand map combines the production nodes on the Y-axis to identify and depict the changes and distributions of resource requirements in each stage of the project life cycle. The supply side is represented by the Z-axis, showing the company's existing and potential digital product resources. By sorting out the digital products on the Z-axis, a product map is formed to clarify which products can meet the various requirements on the demand side. On this basis, a value map is further formed in combination with the product map to show how digital products create value in different demand scenarios, so as to achieve the maximum utilization of resources.

[0042] Refer to Figure 3 , as a further implementation method of the digital resource sharing method, before the step of establishing the pedigree relationship between resources and business scenarios based on the three-dimensional mapping model, it also includes: Step S201, constructing a functional module theme library based on business scenario tags and associating the resources in the structured resource pool with the corresponding functional modules; Among them, the functional module theme library includes but is not limited to a resource intelligent search module, a digital activity module, a product sharing module, a digital course module, a communication community module, and a map dictionary module; Specifically, the business scenario tag is a concrete carrier of the domain ontology. For example, the "bridge health monitoring" scenario tag is mapped to a set of functional requirement sets (including monitoring data call specifications, expert consultation protocols, etc.). When constructing the theme library, the discrete resources in the structured resource pool can be forced to be classified through a semantic constraint binding algorithm.

[0043] Exemplarily, when there is an intersection between the resource attributes and the module definitions (such as a certain BIM model simultaneously meeting the version management rules of the "product sharing module" and the association parsing conditions of the "map dictionary module"), the system adopts a multiple instance allocation strategy based on a rule engine. For example, the resource ID of the "tunnel settlement prediction algorithm" is simultaneously associated with the product sharing module (as a downloadable tool) and the map dictionary module (as a topological node). The resource copies share physical data blocks through a pointer reuse mechanism in the storage layer, avoiding redundant storage. The functional module theme library is essentially a virtual resource container, and its boundary is dynamically defined by an access control list (ACL): only users with discussion permissions are allowed to write resource comments in the communication community module, while the product sharing module is open to authenticated users for download channels. This design enables multi-dimensional logical segmentation of resources in a unified physical storage pool.

[0044] It is understandable that enterprise digital resources are upgraded from "data objects" to "service subjects". Taking the practice of a port group as an example: when a user enters the "intelligent terminal" business scenario, the theme library automatically activates the associated modules - the product sharing module pushes the loading and unloading scheduling algorithm toolkit, the communication community module loads the port automation discussion area, and the graph dictionary module visualizes the topological relationship between algorithms and devices. The resource call response time is shortened to the millisecond level, and due to the modular isolation mechanism, the "port safety specifications" file in the knowledge base will not flow into the digital course module by mistake.

[0045] Step S202: Based on the user query logs of the resource intelligent search module, use a natural language processing model to output a query intent vector set as the input feature of the business process requirement graph and the production stage requirement graph; Specifically, this step is a latent semantic modeling process of human-computer interaction intent.

[0046] Exemplarily, according to the original statement in the user query log (such as "obtain coastal soft foundation settlement cases"), first eliminate semantic noise through the word space projection algorithm: delete stop words ("of", "obtain"), standardize professional terms ("soft foundation" unified as "soft soil foundation"), and complete abbreviations ("BIM" expanded to "Building Information Modeling"). Subsequently, use a fine-tuned BERT model to perform intent distillation: after the model is trained on vertical domain corpora (historical case reports, engineering standard texts), encode the query statement into a 128-dimensional vector. Queries with similar semantics in the vector space are automatically clustered. The cosine similarity between the vectors of "soft soil foundation treatment solutions" and "coastal roadbed reinforcement cases" reaches 0.93, while that of "bridge crack repair" is only 0.21. This vector set serves as the driving factor for the business process requirement graph (X-axis) - when the business process node loads the "soft soil foundation construction" stage, the graph automatically retrieves resource nodes with a similarity > 0.8 and injects them into the execution chain.

[0047] Step S203: Calculate the heat score vector based on the resource download volume and user ratings of the product sharing module as the input feature of the product ability matching graph; Among them, the heat score is a resource value quantification system guided by collective wisdom. The resource download volume d r represents the demand intensity, and the user rating S r reflects the quality recognition. The two are fused through a weighted harmonic function . The logarithmic function compresses the influence of the extreme value of the download volume (to avoid short-term sudden downloads distorting the long-term value), and the rating coefficient focuses on the user's subjective evaluation. Finally, the generated heat vector H=(h r1 ,h r2 ,...,h rm)Input the product capability matching map (Z-axis) to trigger the resource sorting recalibration mechanism. For example, the initial score s of a certain tunnel monitoring software r = 4.2. After a major tunnel accident, the single-week download volume soars to d r = 152, h r jumps to 9.7 points, making it surpass the original top tool in the Z-axis sorting. This is essentially a dynamic anti-entropy process that breaks the Matthew effect in the resource cold start stage, enabling newly launched high-quality tools (such as AI-based settlement prediction models) to quickly enter the top of the recommendations.

[0048] Step S204: Optimize the edge weights of the resource relationship topology graph output by the graph dictionary module to generate a topological adjacency matrix as the input feature of the three-dimensional mapping model.

[0049] Among them, edge weight optimization is a resource relationship purification process based on graph structure entropy. The edge weight w of the original resource relationship topology graph ij reflects the resource association strength (e.g., w = 0.83 between "Shield Machine Parameter Manual" and "Earth Pressure Balance Calculation Tool"), but there are two types of noises: low-weight edges (w < 0.3) are mostly accidental associations (such as false connections generated by a certain user downloading unrelated resources at the same time), and high-weight edges (w > 0.9) may have over-coupling (such as repeated associations of different versions of the same case). The optimizer uses double-threshold Gaussian filtering: perform a shear transformation (w new = 0) on the edges with w < 0.3, and perform Gaussian enhancement on the edges with w > 0.7. The finally generated topological adjacency matrix suppresses topological noise and strengthens the key path at the same time. This topological adjacency matrix is used as the skeleton input of the three-dimensional mapping model, ensuring the path accuracy during the pedigree relationship modeling.

[0050] In the above implementation, the resource service is realized by encapsulating the virtual container of the theme library, the fuzzy business requirements are transformed into graph-driven factors by using the intention vector space mapping, an innovative heat score dynamic anti-entropy mechanism is designed to break the resource cold start dilemma, and the input of the pedigree modeling is purified by relying on the double-threshold optimization of the topological adjacency matrix, and finally a closed-loop technology chain of "business requirements → resource activation → value quantification → relationship purification → pedigree fusion" is formed.

[0051] Refer to Figure 4 , as an implementation of step S104, the steps of establishing the pedigree relationship between resources and business scenarios based on the three-dimensional mapping model, and generating the business process requirement map, production stage requirement map, and product capability matching map respectively include: Step S301: Identify the dependency relationships of the business process nodes in the first dimension, associate the business process node sequences in the first dimension, output the business process node chain, and obtain the business process requirement map covering the entire business process; Specifically, by analyzing the dependency relationships of business process nodes, a complete business process requirements graph is constructed. A business process node refers to the smallest unit of a specific business activity in an enterprise. For example, in the construction industry, "design review" or "construction drawing review" can both be regarded as a node. Dependency relationship means the logical sequence or causal connection between nodes. For instance, the "design review" node must be completed before the "construction drawing review" node, otherwise it will lead to resource scheduling conflicts.

[0052] When identifying these relationships, the system can adopt graph theory algorithms (such as depth-first search or adjacency matrix calculation) to automatically traverse all nodes, detect the input-output dependencies between nodes (such as the output of node A is the input of node B), and ensure that the sequence is acyclic and covers the entire process. The process of associating node sequences is to concatenate the identified nodes in chronological or logical order to form a business process chain. For example, a linear sequence from "requirement collection" to "solution design" to "implementation execution". When outputting, the system converts this chain into a business process requirements graph.

[0053] It should be noted that the business process requirements graph is a graphical data structure, using nodes to represent business activities and edges to represent dependency relationships. It is similar to a flowchart but emphasizes more on resource requirement mapping (such as each node is marked with the required resource type). The construction of the graph utilizes knowledge graph technology to visualize abstract business processes, ensuring that even without looking at the original data, one can intuitively understand how resources support the business flow. The entire process depends on the training of enterprise historical data, and machine learning models are used to predict unknown dependencies to avoid errors caused by manual intervention. This business process requirements graph can automatically identify resource gaps (such as the lack of specific tools in the construction stage), reduce the time of manual inspection, and at the same time ensure that resource requirements are aligned with the enterprise strategy (such as preferentially matching high-value business nodes), avoiding resource waste, and significantly improving the accuracy and efficiency of business resource scheduling.

[0054] Exemplarily, taking a highway construction project as an example: The system identifies the business process node sequence as "geological exploration → design plan formulation → construction bidding → on-site construction → acceptance and delivery". Dependency relationship analysis shows that "design plan formulation" depends on the data output of "geological exploration". If this step is skipped, the "construction bidding" node cannot be started. After associating the sequences, a business process chain is output, and finally a graph is generated: In the graph, the "geological exploration" node points to the "best exploration case" in the benchmark library, while the "on-site construction" node is associated with the "safety specification document" in the knowledge base, covering the entire process from planning to delivery. If a certain project skips "acceptance and delivery", the graph will automatically give an early warning that the resources are not closed-loop.

[0055] Step S302, associate the production life cycle node sequence in the second dimension, output the production stage resource requirement matrix, and obtain the production stage requirement graph; Among them, the production life cycle nodes include the planning and design, construction, and operation and maintenance stages; this step focuses on the time dimension modeling of the production life cycle, and constructs a production stage requirement map by associating node sequences. The production life cycle nodes represent the full cycle stages of a product from concept to abandonment, including planning and design (such as conceptual design and feasibility analysis), construction (such as material procurement and on-site construction), and operation and maintenance (such as equipment maintenance and performance monitoring). These node sequences are not simple lists, but dynamic associations based on time series analysis and state machine models.

[0056] Specifically, the system first defines the standard attributes of each stage (for example, the planning and design stage requires "innovative resources", and the construction stage requires "executive resources"), and then uses matrix operations (such as determinant transformation) to associate the nodes in the order of the life cycle (such as planning and design → construction → operation and maintenance). Finally, a production stage resource requirement matrix is output, which is a two-dimensional table or data grid. The rows represent resource types (such as manpower, equipment, knowledge), the columns represent life cycle stages, and the filled values in the cells represent the intensity of resource requirements (such as a weight value from 0 to 1).

[0057] It should be noted that when this matrix is transformed into a production stage requirement map, visualization techniques (such as heat maps or node-link diagrams) are used. Each stage node in the map is bound to a resource requirement value, allowing users to interactively query (such as hovering to view the detailed resource list for the construction stage). In principle, the generation of the matrix depends on enterprise operation data (such as historical project records), and regression analysis is used to predict resource requirement changes to ensure that the model adapts to different project scales.

[0058] Exemplarily, in the construction of a smart port, the production life cycle sequence is "planning and design (dock layout design) → construction (crane installation) → operation and maintenance (automation system monitoring)". The system associates the sequence and outputs a matrix: in the matrix, the resource requirement value of "crane" in the construction stage is 0.9 (high demand), and the requirement value of "AI monitoring tool" in the operation and maintenance stage is 0.7. The final requirement map is displayed as a heat map, in which the "construction" node is highlighted, linked to the "BIM modeling software" in the product library, and warns that if there is a shortage of resources in the planning and design stage (such as a shortage of design software), it will cause downstream delays.

[0059] Step S303, associate the digital product attribute sequence in the third dimension, and output a product ability score table by binding the matching weight value of the digital product and the business scenario, to obtain a product ability matching map.

[0060] Among them, the core of this step is the value quantification and scenario adaptation of digital products, and a product capability matching map is generated through the association of attribute sequences. The digital product attribute sequence refers to the sequence of key feature dimensions of the product (such as functionality, compatibility, cost-effectiveness), for example, the "processing speed" or "user-friendliness" of software tools.

[0061] Specifically, the system first defines a standard framework for the attribute sequence (based on industry norms), and uses multidimensional scaling analysis (MDS) or principal component analysis (PCA) to map the attributes to a numerical space (such as quantifying "processing speed" from 1 to 10 points). When binding the matching weight values, a weighted fusion algorithm is adopted: calculate the matching weights between the digital product (such as an AI analysis tool) and the business scenario (such as bridge monitoring), and the weight values are dynamically calculated based on historical usage data (download volume, user ratings) and scenario requirements (urgency, complexity). For example, the weight formula is: weight = functionality score × 0.4 + compatibility score × 0.3 + cost score × 0.3). The output product capability score table is a structured data set (such as in JSON or CSV format), listing the attribute scores and comprehensive weights of each product. Finally, it is converted into a product capability matching map, where the nodes in the map represent products, the edges represent the matching relationships with business scenarios, and the edge weight values visualize the matching intensity (such as the thickness of the lines). The entire process integrates machine learning (such as clustering algorithms) to automatically optimize the weight allocation and ensure that the map reflects real-time business needs.

[0062] Exemplarily, in a railway construction project, the digital product sequence includes "track fine-tuning software (high functionality)", "settlement prediction model (medium compatibility)", and "BIM collaboration platform (low cost)". When binding the matching weights, the weight of the software with the "track laying" scenario is 0.85 (because the functionality score is 9 / 10), and the weight of the model with the "geological monitoring" scenario is 0.6. After the score table is output, the map shows that the "track fine-tuning software" node is connected to the "track laying" scenario node with a thick edge, indicating that this resource should be deployed preferentially; if the cost attribute changes, the map will update the weights in real time.

[0063] In the above embodiments, the business process requirement map (the first dimension) ensures full-process coverage of resources and avoids business interruptions; the production stage requirement map (the second dimension) optimizes the life-cycle resource allocation and reduces waste; the product capability matching map (the third dimension) realizes precise dynamic matching and improves efficiency. By upgrading resource management from static storage to intelligent drive, the resource invocation time is shortened, the matching accuracy is improved, and strategic decision-making is supported through map visualization, thus providing a core engine for the digital transformation of enterprises.

[0064] It should be noted that the three-dimensional pedigree mapping mechanism in the embodiments of this application is different from the planar classification of traditional resource management systems. This mechanism establishes a multi-level mapping of resources and services from the spatial dimension. The business process dimension (X-axis) reveals the positioning of resources in specific operation chains, the production life cycle dimension (Y-axis) reflects the evolutionary trajectory of resources in the time flow, and the digital product dimension (Z-axis) quantifies the supply capacity characteristics of resources. This three-dimensional coupling model enables enterprises to see through the value contribution rate of resources in specific business scenarios. For example, in the intelligent construction scenario of highways, by positioning BIM technology nodes through the X-axis, associating construction stages through the Y-axis, and invoking model library resources through the Z-axis, an accurate technical solution can be finally formed.

[0065] Referring to Figure 5 , as an implementation manner of step S105, the steps of constructing a visual digital resource capability platform include: Step 401, construct a four-level map framework, which includes a full-industry panoramic map layer, a single-industry professional map layer, a single-stage resource map layer, and an application scenario card layer; Specifically, by defining a four-layer progressive topological structure, a panoramic modeling system of enterprise resources is established. The full-industry panoramic map layer adopts a directed weighted graph model (nodes represent industrial sectors, and edge weights reflect the cross-industry resource dependence intensity), mapping the global business chain and resource allocation relationship from the group strategic height. The single-industry professional map layer is based on a life cycle state machine (such as planning and design → construction → operation and maintenance), decomposing the full process of a single industry into discrete state nodes, and binding a standardized resource demand template for each node at this stage. The single-stage resource map layer introduces a process-level knowledge graph, extracts process nodes, resource entities and their association rules in specific production scenarios (such as bridge pile foundation construction) through entity recognition algorithms, and forms a fine-grained matching path. The application scenario card layer constructs a scenario template instantiation engine, converts unstructured requirements such as business pain points and resource constraints into a standardized field combination (such as "process type: pile foundation inspection; matching tool: ultrasonic flaw detector"), and realizes the product-level dimensionality reduction mapping of high-dimensional business elements to digital resources. The four levels ensure data logic consistency through a parent-child inheritance relationship (such as the "infrastructure" node in the full-industry layer is associated with the "highway construction" sub-graph in the single-industry layer), forming a complete decision-making chain from macro strategy to micro execution.

[0066] It can be understood that this four-level map architecture solves the problem of disconnection between strategy and execution in traditional resource platforms, enabling enterprises to penetrate layer by layer from the group perspective to the resource allocation plan for specific processes. The four-layer structure provides a standardized container for subsequent three-dimensional data fusion, avoids cross-departmental data redundancy, and at the same time the standardized template reduces the user's understanding cost (such as the scene card fields can be directly parsed by business personnel).

[0067] Step 402, inject the business process node chain into the single-stage resource map layer; Among them, when the business process node chain (X-axis data) is injected into the single-stage resource atlas layer, the key steps in the business chain (such as "construction drawing review") are strongly associated with the entity nodes in the process atlas through the node alignment algorithm. The core is to calculate the semantic similarity score (such as the node similarity between "construction drawing review → design verification" is 0.92), and establish a two-way index for nodes with a score > 0.8.

[0068] Step 403, inject the production stage demand matrix into the single-industry professional atlas layer; Among them, when the production stage demand matrix (Y-axis data) is injected into the single-industry professional atlas layer, the resource demand characteristics in the matrix (such as "temperature and humidity monitoring accuracy of ±0.5°C is required during the concrete curing stage") are converted into the attribute values of the graph nodes, driving the resource adaptation engine to automatically match the candidate solutions in the knowledge base (such as matching "intelligent temperature control sensor - type A").

[0069] Step 404, bind the digital product attribute sequence to the nodes of the full-industry panoramic atlas layer; Among them, when the digital product attribute sequence (Z-axis data) is bound to the nodes of the full-industry panoramic atlas layer, the attribute fusion technology is used to compare the product technical parameters (such as response delay ≤ 50ms) with the business requirement indicators of the industrial nodes (such as "real-time monitoring scenario requires a delay ≤ 100ms"). If it is satisfied, a binding relationship is established and written into the edge attribute of the atlas (the edge type is marked as "technical support").

[0070] It can be understood that through the three-dimensional data injection, the precise anchoring of business requirements and product capabilities is achieved at the resource logic layer. For example, the X-axis business flow ensures that resource scheduling complies with the engineering process specifications, the Y-axis demand characteristics guarantee that the resource technical parameters meet the production constraints, and the Z-axis product binding opens up the technology supply channel. The three form a closed-loop decision-making basis in the atlas.

[0071] Step 405, call the weight value data of the product ability matching atlas, and fill the scene card fields in the application scenario card layer based on the industry code of the single-industry professional atlas layer, the process node identifier of the single-stage resource atlas layer, and the industry sector node of the full-industry panoramic atlas layer; Among them, call the weight value data in the product ability matching atlas (such as resource popularity score = download frequency × 0.6 + user score × 0.4), and combine the industry code of the single-industry professional atlas layer (such as "highway engineering → LY-2023" indicating the road type and version), the process node identifier of the single-stage resource atlas layer (such as "pile foundation detection → ZJ-008"), and the industry sector node of the full-industry panoramic atlas layer to construct multi-dimensional retrieval conditions. The weight threshold screening module generates a high-confidence resource list through a dynamic filtering algorithm (such as only selecting the top 20% of products with the highest weight values).

[0072] Next, the scenario card filling engine converts the above data into business-readable fields (such as "Recommended Product: Ultrasonic Flaw Detector - UltrasoundPro; Matching Degree: 92%; Applicable Process: ZJ-008") based on the field mapping rule library and writes them into the predefined field slots of the card. During the card generation process, the superior graph relationship is automatically inherited (for example, a "Bridge Pile Foundation Detection" scenario card is associated with the single-industry layer "Highway Engineering" and the entire-industry layer "Infrastructure Sector"), forming cross-level traceability.

[0073] Step 406: Deploy a graph dynamic update engine to monitor the new resource stream of the data access interface and trigger hierarchical updates of the graph according to the new resource type. Among them, deploy an event listener to capture the new data stream of the API interface in real time (such as adding a new product "5G Intelligent Monitoring Reagent"), and identify the data type through a resource classifier (such as marking product attribute changes as Z-axis updates).

[0074] Specifically, the engine triggers hierarchical incremental updates according to the data type. For example: for business process changes (X-axis), use the sub-graph recalculation algorithm to locally refresh the affected nodes in the single-stage resource graph layer (such as "Construction Process Optimization" causing related process nodes to be updated); for production demand changes (Y-axis), trigger demand matrix iteration in the single-industry professional graph layer (such as expanding the matrix dimension when adding an "Environmental Protection Requirement" indicator); for product data changes (Z-axis), perform a re-evaluation of the binding relationship in the entire-industry layer (such as automatically creating a binding edge between nodes when the new product parameters meet existing requirements).

[0075] It should be noted that the update process adopts an impact propagation constraint mechanism (such as defining the node impact radius ≤ 3 hops), and only refreshes the relevant sub-graphs to reduce the calculation overhead. The update results are synchronized to the scenario card layer in real time, driving dynamic refreshing of the card fields (such as automatically adding new products to the recommendation list when they meet the threshold).

[0076] Step 407: Integrate the four-level graph framework and the graph dynamic update engine into the visual interaction interface to build a visual digital resource capability platform.

[0077] Specifically, encapsulate the four-level graph framework and the dynamic engine into a microservice architecture, and connect to modules such as resource retrieval and permission control through a unified data bus. The visual interaction interface adopts a hierarchical rendering technology: the entire-industry layer presents an industry sector heat map, the single-industry layer shows the life cycle time axis, the single-stage layer generates a process node topology map, and the scenario card layer arranges in a card matrix. User operations (such as clicking on the "Highway Construction" node) trigger cross-level linkage responses (automatically drilling down to all process cards under this industry). The platform integrates a permission inheritance model (such as group users can view the entire-industry layer, and project departments can only see the associated single-stage layer) to ensure data security and business focus.

[0078] In the above embodiments, a knowledge ontology of enterprise resources is constructed based on a four - level framework. The injection of three - dimensional data realizes the accurate transformation from business requirements to technical solutions. The encapsulation of scenario cards provides a resource solution that is ready - to - use for the business side. The dynamic update engine ensures that the system continuously responds to business changes. Finally, the platform integrates the scattered digital resources of the enterprise into a decision - making system that runs through strategy - tactics - execution in a visual form, greatly reducing the resource retrieval cost, improving the cross - departmental resource reuse rate, and enhancing the long - term adaptability of the platform through a dynamic mechanism.

[0079] Referring to Figure 6 , as an embodiment of step S405, the steps of calling the weight value data of the product - capacity matching map and filling the scenario card fields in the application scenario card layer based on the industry code in the single - industry professional map layer, the process node identifier in the single - stage resource map layer, and the industry sector node in the full - industry panoramic map layer include: Step S501, call the screening result of the weight threshold of the product - capacity matching map, filter the candidate products with weight values lower than the preset weight threshold, and generate a list of filtered candidate products; Among them, the product - capacity matching map is a dynamic database for quantitatively evaluating product value. The weight - threshold screening depends on a preset algorithm model (such as weight value = download frequency × α+user rating × β). The essence of the preset weight threshold is a resource - value filtering line. For example, only products with a weight > 85 points are selected, dynamically intercepting low - efficiency resources (such as unpopular tools or unproven new products), and generating a list of filtered candidate products (a subset of high - value resources). The essence of this step is a resource - value distillation mechanism.

[0080] Step S502, receive the input of the industry code corresponding to the single - industry professional map layer; Step S503, based on the industry code, retrieve the resource adaptation rule library corresponding to the industry code; Among them, the industry code corresponding to the single - industry professional map layer is the digital identifier of the industry divided within the enterprise. The essence of this industry code is a business semantic label, which binds the resource - usage rules unique to this industry (such as the highway engineering rules require that the card must contain the field of "pile foundation deformation compensation algorithm"). The system retrieves the resource adaptation rule library (a resource - configuration logic library stored by industry classification, for example, the highway engineering rule library mandates that "pile foundation detection needs to be associated with an error - correction algorithm"), which is actually mapping through the code to the resource - scheduling protocol of the corresponding industry.

[0081] It can be understood that by transforming abstract business requirements into machine - recognizable configuration instructions, it prevents the misuse of cross - industry resources (such as misconfiguring water - conservancy engineering tools to highway scenarios).

[0082] Step S504: Generate solution cards based on the filtered candidate product list and the resource adaptation rule library. Among them, the resource adaptation rule library defines the generation logic of the card content. After the system parses the rules, it extracts compliant product instances from the candidate product list according to strong constraint conditions and assembles them into a solution card prototype (semi-structured data container).

[0083] Step S505: Fill the solution cards into the predefined field slots of the application scenario card layer to form structured card instances. Among them, the predefined field slots of the application scenario card layer are essentially standardized templates. The system fills the solution card prototype according to the slot semantics to form structured card instances (machine-readable complete business objects). This process depends on the field semantic alignment algorithm to ensure the consistency between business terms and template definitions (e.g., "error correction" corresponds to the standardized field "process optimization plan").

[0084] Step S506: Associate the process node identifiers in the single-stage resource map layer with the life cycle stage codes in the single-industry professional map layer. Among them, the process node identifier (such as "ZJ-008") is essentially a digital label of the business atomic unit, and its coding rule implies business attributes (e.g., the first two letters "ZJ" represent pile foundation detection, and the digital segment "008" identifies the specific detection point). The system disassembles the syntax structure of the identifier (such as delimiters, prefix codes) through the business semantic parsing engine, extracts key business features (process type, physical location, etc.), and matches them with the life cycle model in the single-industry professional map layer. The association logic depends on the decision tree rule library. For example, if the process node identifier contains the semantic of "pile foundation detection" (determined by NLP keyword clustering), it is forced to be associated with the "construction stage" code; if the process contains the feature of "scheme comparison", it is associated with the "design stage" code.

[0085] Step S507: Associate the industrial sector nodes in the full-industry panoramic map layer through the industrial code. Among them, the industrial sector nodes in the full-industry panoramic map layer are the root nodes of the group-level business classification tree, which store the metadata of all subordinate industries (including the industrial code mapping table). The system uses the industrial code as the key value to query the distributed hash index of the panoramic map to locate the target industrial sector node. Step S508: Write the hierarchical traceability identifier into the structured card instance. Among them, the hierarchical traceability identifier is generated based on the industrial sector node and the life cycle stage code.

[0086] Specifically, the format of the hierarchical traceability identifier is a concatenated sequence of industry sector ID, industry ID and process node ID. This identifier is the unique grid coordinate of the resource in the panoramic business map.

[0087] It is understandable that the application scenario card layer breaks the data barriers at the business level, allowing any scenario card to be traced back to the group strategy layer (panoramic map), business execution layer (single industry map), and process node layer (single stage map), realizing end-to-end transparent management of resource application.

[0088] In the above implementation, from industry coding to rule library binding, a business-oriented resource screening framework is established, high-value resources are screened using a weight threshold mechanism, and standardized solution cards are generated through a rule-product dual-driven model. Ultimately, global traceability of resource applications is achieved based on a three-dimensional cascade coordinate system (full industry layer / single industry layer / single stage layer).

[0089] Reference Figure 7 As an implementation of step S106, in response to the resource call instruction input by the user, the step of outputting the target resource identifier and the associated path based on the visual digital resource capability platform includes: Step S601: Parse the semantics of the resource call instruction and identify the target resource type and business domain according to the preset resource classification standard; Specifically, it relies on the named entity recognition and part-of-speech tagging mechanism in natural language processing (NLP) technology to identify entity objects and their contextual business context in user input. Its essence is the translation process from natural language to digital business space.

[0090] For example, when the user inputs a command such as "coastal soft foundation treatment case", the semantic parsing engine breaks down the sentence structure through a bidirectional LSTM model: "coastal" → geographical attribute label → mapping water transport field code; "soft foundation treatment" → technical action label → associated foundation treatment node; "case" → resource type label → pointing to the benchmark library.

[0091] Next, a pre-set resource classification standard serves as a translation dictionary, transforming discrete terms into structured business coordinates. The system categorizes user-required resources into standard resource types and further maps them to their respective business domains (such as smart manufacturing or intelligent transportation), thereby determining the semantic boundaries and business objectives of resource invocation.

[0092] Step S602: determining the three-dimensional coordinates of the corresponding target resource node based on the single industry professional map; Among them, the three-dimensional coordinates include the axis position of the business process, the axis position of the production node and the axis position of the resource type. Its essence is the unique matrix address of the business scenario in the digital space, which solves the resource mismatch problem caused by semantic ambiguity in traditional retrieval.

[0093] Specifically, after the basic semantics of the resources, the system will further perform spatial positioning of the resource nodes through the constructed single-industry professional map. This map is constructed in a three-dimensional coordinate system, using the X-axis to represent the business process stage, the Y-axis to represent the production node location, and the Z-axis to represent the resource type dimension. In this model, the business process axis code (X-axis) is usually encoded according to the entire industrial life cycle (such as design, R & D, production, operation and maintenance, etc.), the production node code (Y-axis) reflects the organizational relationship of different process nodes or application scenarios within the industry, and the resource type axis (Z-axis) reflects the technical attributes of the resources, such as document type, algorithm type, tool type, etc. By mapping the target resource nodes into this three-dimensional coordinate space, the system realizes the semantic positioning of the resource space, thus having the structural basis for constructing the paths and context associations between resources.

[0094] Step S603: Extract all the associated edges of the target resource nodes from the map dictionary, calculate the weight value of each associated edge, filter the associated edges lower than the preset threshold, and generate a set of high-weight associated paths. Specifically, the edges in the map represent the semantic or operation path relationships between resources. The weight of the associated edges can be calculated through a dynamic statistical model, and its calculation formula is: weight = call frequency × α + user score × β. The call frequency represents the number of times this resource edge has been actually accessed in history, reflecting the operation activity; the user score comes from the user feedback mechanism and is used to measure the satisfaction and effect of this resource path. α and β are empirical parameters, generally satisfying α + β = 1, and their values can be dynamically adjusted through methods such as A / B testing to adapt to different business scenarios.

[0095] This calculation method combines the analysis of behavior frequency and the evaluation of user subjective experience, ensuring that the recommended paths are both operationally representative and can reflect user preferences. After the calculation, the system sorts all the edges according to the weight and filters out the low-quality paths according to the set threshold (such as the weight value is greater than 0.6), and finally obtains a candidate set of resource paths with high correlation.

[0096] Step S604: Sort the set of high-weight associated paths according to the total weight, generate corresponding hierarchical traceability identifiers for the resource nodes in each path, and output a resource path object including the hierarchical traceability identifiers. Among them, the hierarchical traceability identifier includes the industrial sector code, industrial code, production node code, and resource node code. Specifically, obtain the industrial sector code from the panoramic map of the entire industry, obtain the industrial code and production node code from the professional map of a single industry, and obtain the resource node code from the resource node metadata; the industrial sector code is obtained from the topological tree at the entire industry level. For example, "coastal infrastructure" → COASTAL-INFRA, representing the resource strategic level attribution; the industrial code is extracted from the single industry level. For example, "water transportation project" → SW-2023, identifying the business execution domain; the production node code inherits the Y-axis coordinate to locate the process link; the resource node code is assigned a unique ID by the resource type library.

[0097] It should be noted that this hierarchical coding structure can not only achieve reversible retrieval and structured management of resource paths, but also enhance the interoperability of resources between heterogeneous systems, providing a reliable identity reference mechanism for data sharing and auditing. The path set and its corresponding identifier will then be output as a resource path object to the user interface for the user to select.

[0098] Step S605, update the graph weight data according to the resource path object selected by the user, adjust the call frequency and user score of the associated edge, and trigger the dynamic update of the topological relationship of the graph dictionary.

[0099] Among them, when the user completes the adoption and call of the resource path, the system needs to update the edge weight in the graph structure in real time to maintain the dynamic evolution and self-optimization of the graph.

[0100] Exemplarily, this process first updates the call frequency of the edge incrementally through the call log, and then adjusts the user score weight value in a weighted average manner according to the feedback score provided by the user and the historical score record, so as to update the overall weight of the edge. All weight updates will be written to the persistent layer of the graph dictionary (such as the graph database Neo4j or the GraphQL storage engine), and trigger the data synchronization update of the visualization layer. In addition, when the call frequency of a certain path continuously exceeds the preset threshold (such as the call count is greater than 50), the system will automatically mark it as a "recommended link" and push it to the popular resource recommendation module to improve resource utilization efficiency and guide user behavior.

[0101] In the above embodiment, the resource call method based on pedigree mapping can not only realize the full process automation from the user's natural language input to the accurate identification and path recommendation of resources, but also construct a dynamically self-evolving resource graph structure. The three-dimensional coordinate system enables resource positioning to have double constraints of business process and technical semantics. The high-weight path calculation integrates behavioral data and subjective evaluation to improve the rationality of recommendation, while the hierarchical traceability identifier enhances the interpretability and interoperability of resource paths.

[0102] Refer to Figure 8 , as a further embodiment of the resource sharing method, it further includes: Step S701: Invoke the resource demand intensity data of each node in the business process requirement graph and the resource supply intensity data of the corresponding nodes in the product capability matching graph. Among them, a quantitative mapping channel of "business demand - resource supply" is established. When the system invokes the resource demand intensity data of a certain node (such as the "pile foundation construction process") in the business process requirement graph (X-axis), it is actually extracting the quantitative value of the resource urgency of this node in the business scenario. At the same time, the resource supply intensity data (such as the supply intensity of the "intelligent pile driving monitoring SDK") corresponding to the coordinate of the product capability matching graph (Z-axis) is invoked. This value is derived from the resource pool call records and performance evaluation indicators. Transforming the abstract business scenario (pile foundation construction) and physical resources (monitoring SDK) into computable numerical pairs provides standardized input for subsequent value evaluation.

[0103] Step S702: Calculate the overlap coefficient based on the coincidence degree of the resource demand intensity data and the resource supply intensity data, and generate a two-dimensional resource value matrix. Among them, the system calculates the overlap degree of resource supply and demand matching by invoking the supply and demand intensity data of the graph nodes. In the business process graph, the "resource demand intensity" marked for each node can be quantified as a demand index (such as Q demand = 10 unit resources / hour), while in the product capability matching graph, the "supply intensity" of the resource node at the same coordinate can be expressed as the current available resource service capacity (such as Q supply = 7 unit resources / hour).

[0104] Specifically, the overlap coefficient calculation method proposed in the embodiment of the present application is to take the smaller value of the two and multiply it by the value weight factor configured in the business domain. The value weight factor reflects the industry priority (such as a higher coefficient for key process points) and is configured by inheriting domain knowledge from the professional graph. For example, when Q demand = 10, Q supply = 7, and the weight is 0.9 for a certain node, the overlap coefficient is 6.3. Filling the overlap coefficients of all node pairs into the two-dimensional matrix generates the resource value matrix. This matrix takes the business process node and the resource type coordinate as dimensions, and the matrix value reflects the resource supply and demand fit degree at this coordinate position.

[0105] Step S703: Perform a time-axis alignment mapping on the time sequence nodes of the production stage demand graph and the process nodes of the business process demand graph to generate a spatio-temporal association matrix. Specifically, after spatial analysis, to consider the association between tasks and production line time, this method also performs a time-axis alignment mapping between the process nodes in the business process requirement graph and the time-series nodes in the production stage requirement graph. The business nodes (X-axis) are aligned and mapped to the corresponding production time-series nodes (Y-axis) to form a "spatial-temporal association matrix". The rows of the matrix represent the Y-axis time-series nodes, the columns represent the X-axis business process nodes, and the matrix cell values are the total resource requirements of the corresponding process nodes at this time series. This mapping ensures that resource matching analysis occurs not only in the spatial dimension but also considers the actual demand evolution in the time dimension, thereby enhancing the timeliness of the analysis and the task matching accuracy.

[0106] Step S704, scan the coordinate regions in the two-dimensional resource value matrix with an overlap coefficient lower than the preset overlap threshold, and combine the total resource data of the corresponding nodes in the spatial-temporal association matrix to generate a resource gap warning signal.

[0107] Among them, the system scans the low-matching regions in the two-dimensional resource value matrix to identify the positions where the resource supply-demand overlap coefficient is lower than the preset overlap threshold. For example, if the overlap coefficient of a certain coordinate is less than 0.3, it indicates that the resources are extremely scarce. At this time, the total resource demand value corresponding to this coordinate is extracted from the spatial-temporal association matrix and compared with the actual supply quantity in the product capability matching graph. Based on the gap degree calculation formula: (total resource demand - actual supply quantity) / total resource demand, the system calculates the resource gap degree. When the gap degree is greater than the warning threshold (such as 60%), the system generates a warning signal and indicates the specific location of the gap (i.e., the time-series node, process node, resource type) and the resource identifier. This resource identifier is a hierarchical classification code and can be directly used to point to the corresponding resource entry in the resource pool (such as project documents or code packages in the benchmark library or product library).

[0108] It should be noted that the warning signal not only has a notification function but also can trigger an automatic response mechanism. The system calls the corresponding resources in the resource pool according to the identifier to fill the corresponding nodes in the supply graph and update their supply intensity data. At the same time, the cell value at this position in the resource value matrix is corrected to achieve closed-loop control. In this way, the resource sharing platform has the capabilities of dynamic allocation, predicting shortages, and automatic adjustment, significantly improving the efficiency and response ability of enterprise resource utilization.

[0109] In the above embodiments, a multi-dimensional resource analysis mechanism that combines the construction of a three-dimensional map with a two-dimensional resource value matrix and a spatio-temporal correlation matrix is established, realizing the fine characterization and dynamic identification of the resource supply and demand status. Based on the pedigree mapping relationship among business process requirements, production stage rhythms, and resource capabilities, the system can identify resource shortages in advance, automatically activate the resource pool through warning signals, and dynamically repair the imbalance between resource supply and demand. The embodiments of the present application achieve a fundamental transformation of the resource sharing system from static supply to intelligent allocation and from manual judgment to data-driven evolution, with high industrial adaptability, real-time performance, and self-adaptive capabilities, providing basic support for digital factories and intelligent enterprises.

[0110] The embodiments of the present application also disclose a digital resource sharing system based on a pedigree mapping relationship.

[0111] A digital resource sharing system based on a pedigree mapping relationship, the resource sharing system includes: A heterogeneous data acquisition module, configured to obtain multi-source heterogeneous data from the enterprise's distributed system through a preset interface protocol; A data processing module, configured to perform standardized processing on the multi-source heterogeneous data based on a preset resource classification standard to generate a structured resource pool; An associated topology analysis module, configured to build a map dictionary module, perform associated topology analysis on the resources in the structured resource pool, and output a resource relationship topology map; A multi-map generation module, configured to establish a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model according to the structured resource pool and the resource relationship topology map, and respectively generate a business process requirement map, a production stage requirement map, and a product capability matching map; A platform construction module, configured to fuse the business process requirement map, the production stage requirement map, and the product capability matching map to build a visual digital resource capability platform; A target resource identifier output module, configured to respond to a resource call instruction input by a user and output a target resource identifier and an associated path based on the visual digital resource capability platform.

[0112] The digital resource sharing system based on a pedigree mapping relationship in the embodiments of the present application can implement any of the above resource sharing methods, and the specific working processes of each module in the resource sharing system can refer to the corresponding processes in the above method embodiments.

[0113] In several embodiments provided in the present application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the division of a certain module is only a logical function division, and there can be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0114] The embodiments of the present application also disclose a computer device.

[0115] The computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements a digital resource sharing method based on a pedigree mapping relationship as described above.

[0116] The embodiments of the present application also disclose a computer-readable storage medium.

[0117] The computer-readable storage medium stores a computer program that can be loaded and executed by a processor to implement any one of the digital resource sharing methods based on a pedigree mapping relationship as described above.

[0118] Among them, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or component; the program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0119] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] The above are all the preferred embodiments of the present application. Without restricting the protection scope of the present application accordingly, any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A digital resource sharing method based on a pedigree mapping relationship, characterized in that, The resource sharing method includes: Obtaining multi-source heterogeneous data from the enterprise's distributed system through a preset interface protocol; Based on a preset resource classification standard, performing standardization processing on the multi-source heterogeneous data to generate a structured resource pool; Constructing a graph dictionary module to perform associated topology analysis on the resources in the structured resource pool and output a resource relationship topology graph; Based on the structured resource pool and the resource relationship topology graph, establishing a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model, and respectively generating a business process requirement graph, a production stage requirement graph, and a product ability matching graph; Fusing the business process requirement graph, the production stage requirement graph, and the product ability matching graph to construct a visual digital resource ability platform; Responding to a resource invocation instruction input by a user, and outputting a target resource identifier and an associated path based on the visual digital resource ability platform.

2. The digital resource sharing method based on a pedigree mapping relationship according to claim 1, wherein Before the step of establishing a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model, it further includes: Constructing a functional module theme library based on business scenario tags and associating the resources in the structured resource pool to the corresponding functional modules; wherein, the functional module theme library includes a resource intelligent search module, a product sharing module, and a graph dictionary module; Based on the user query logs of the resource intelligent search module, outputting a query intention vector set through a natural language processing model as the input feature of the business process requirement graph and the production stage requirement graph; Calculating a heat score vector based on the resource download volume and user ratings of the product sharing module as the input feature of the product ability matching graph; Performing edge weight optimization on the resource relationship topology graph output by the graph dictionary module to generate a topological adjacency matrix as the input feature of the three-dimensional mapping model.

3. The digital resource sharing method based on the pedigree mapping relationship according to claim 2, characterized in that, The steps of establishing a pedigree relationship between resources and business scenarios based on a three-dimensional mapping model and respectively generating a business process requirement graph, a production stage requirement graph, and a product ability matching graph include: Identifying the dependency relationships of business process nodes in the first dimension, associating a business process node sequence in the first dimension, outputting a business process node chain, and obtaining a business process requirement graph covering the entire business process; Associating a production life cycle node sequence in the second dimension, outputting a production stage resource requirement matrix, and obtaining a production stage requirement graph; the production life cycle nodes include the planning and design, construction, and operation and maintenance stages; Associating a digital product attribute sequence in the third dimension, and outputting a product ability score table by binding the matching weight value between the digital product and the business scenario to obtain a product ability matching graph.

4. A digital resource sharing method based on a pedigree mapping relationship according to claim 3, characterized in that, The steps of constructing a visual digital resource ability platform include: Constructing a four-layer graph framework, including a full-industry panoramic graph layer, a single-industry professional graph layer, a single-stage resource graph layer, and an application scenario card layer; Injecting the business process node chain into the single-stage resource graph layer; Injecting the production stage resource requirement matrix into the single-industry professional graph layer; Binding the digital product attribute sequence to the nodes of the full-industry panoramic graph layer; Call the weight value data of the product capability matching map, and fill the scenario card fields in the application scenario card layer based on the industry code of the single-industry professional map layer, the process node identifier of the single-stage resource map layer, and the industry sector node of the full-industry panoramic map layer; Deploy a map dynamic update engine to monitor the new resource stream of the data access interface, and trigger hierarchical updates of the map according to the new resource type; Integrate the four-layer map framework and the map dynamic update engine into the visual interaction interface to build a visual digital resource capability platform.

5. A digital resource sharing method based on a pedigree mapping relationship according to claim 4, wherein The steps of calling the weight value data of the product capability matching map and filling the scenario card fields in the application scenario card layer based on the industry code of the single-industry professional map layer, the process node identifier of the single-stage resource map layer, and the industry sector node of the full-industry panoramic map layer include: Call the weight threshold screening results of the product capability matching map, filter candidate products with weight values lower than the preset weight threshold, and generate a list of filtered candidate products; Receive the industry code input corresponding to the single-industry professional map layer; Based on the industry code, retrieve the resource adaptation rule library corresponding to the industry code; Generate solution cards based on the filtered candidate product list and the resource adaptation rule library; Fill the solution cards into the predefined field slots of the application scenario card layer to form a structured card instance; Associate the lifecycle stage code of the single-industry professional map layer through the process node identifier of the single-stage resource map layer; Associate the industry sector node of the full-industry panoramic map layer through the industry code; Write a hierarchical traceability identifier into the structured card instance; wherein, the hierarchical traceability identifier is generated based on the industry sector node and the lifecycle stage code.

6. A digital resource sharing method based on pedigree mapping relationship according to claim 4, characterized in that: The steps of responding to the resource call instruction input by the user and outputting the target resource identifier and the associated path based on the visual digital resource capability platform include: Parse the semantics of the resource call instruction, and identify the target resource type and business domain according to the preset resource classification standard; Based on the single-industry professional map, determine the three-dimensional coordinates of the corresponding target resource node; Extract all the associated edges of the target resource node from the map dictionary, calculate the weight value of each associated edge, filter the associated edges lower than the preset threshold, and generate a set of high-weight associated paths; Sort the set of high-weight associated paths by the total weight, generate a corresponding hierarchical traceability identifier for the resource nodes in each path, and output a resource path object including the hierarchical traceability identifier; Update the map weight data according to the resource path object selected by the user, adjust the call frequency and user score of the associated edges, and trigger dynamic updates of the topological relationship of the map dictionary.

7. A digital resource sharing method based on a pedigree mapping relationship according to any one of claims 1 to 6, characterized in that The resource sharing method further includes: Call the resource demand intensity data of each node in the business process requirement map and the resource supply intensity data of the corresponding nodes in the product capability matching map; Calculate the overlap coefficient based on the coincidence degree of the resource demand intensity data and the resource supply intensity data, and generate a two-dimensional resource value matrix; Perform a time-axis alignment mapping on the time sequence nodes of the production stage demand spectrum and the process nodes of the business process demand spectrum to generate a spatio-temporal correlation matrix; Scan the coordinate regions in the two-dimensional resource value matrix with an overlap coefficient lower than the preset overlap threshold, and combine the total resource data of the corresponding nodes in the spatio-temporal correlation matrix to generate a resource gap warning signal.

8. A digital resource sharing system based on a pedigree mapping relationship, characterized in that, The resource sharing system includes: A heterogeneous data acquisition module for acquiring multi-source heterogeneous data from the enterprise internal distributed system through a preset interface protocol; A data processing module for performing standardization processing on the multi-source heterogeneous data based on a preset resource classification standard to generate a structured resource pool; An associated topology analysis module for constructing a spectrum dictionary module, performing associated topology analysis on the resources in the structured resource pool, and outputting a resource relationship topology graph; A multi-spectrum generation module for establishing a pedigree relationship between resources and business scenarios based on the structured resource pool and the resource relationship topology graph, and respectively generating a business process demand spectrum, a production stage demand spectrum, and a product ability matching spectrum; A platform construction module for fusing the business process demand spectrum, the production stage demand spectrum, and the product ability matching spectrum to construct a visual digital resource ability platform; A target resource identifier output module for responding to a resource call instruction input by a user and outputting a target resource identifier and an associated path based on the visual digital resource ability platform.

9. A computer device, characterized in that: It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Stores a computer program that can be loaded and executed by a processor to implement the method according to any one of claims 1 to 7.

Images