Industrial digital management method and system integrating geographical environment data

Abstract

The invention relates to the technical field of big data management, and particularly discloses an industrial digital management method and system integrating geographical environment data, and the method comprises the steps: an industrial digital management platform is hung to a pre-constructed basic management object according to a target time slice, and is written into a corresponding TPE scene unit; under each target time slice, the platform carries out time aggregation on the target setting data, and channel exposure vectors and node structure characteristic parameters are evaluated on an aggregation result; the platform further executes simulation deduction of the integrated geologic environment scene on the target TPE scene unit for predicting the change track of the key operation index; and then obtaining a corresponding industrial input matrix and an industrial output matrix, and jointly configuring a simulation prediction process of the geographical environment scene with simulation input characteristics, thereby completing closed-loop industrial digital management from historical efficiency evaluation to resource and target constraint setting to scene simulation prediction in the same framework.

Description

A digital management method and system for industries integrating geopolitical environmental data Technical Field

[0001] This invention relates to the field of big data management technology, specifically to an industrial digital management method and system that integrates geopolitical environmental data. Background Technology

[0002] Currently, global technological competition has become a core dimension of great power rivalry, and the semiconductor industry, as a key area concerning industrial security and development initiative, has become a strategic focus of global competition. Driven by technological competition among major economies, the global industrial chain is undergoing profound restructuring, and the localization trend in the technology sector is gradually reshaping the global industrial division of labor. Against this backdrop, many countries have incorporated the semiconductor industry into their national strategies, regarding industrial upgrading and breakthroughs as the core direction for enhancing future competitiveness. Some emerging economies face similar external technological constraints and endogenous development demands, yet have chosen differentiated paths to industrial breakthroughs: one type of economy relies on a coordinated and integrated development model to promote the independent layout of each link in the industrial chain, striving to achieve independent control of core technologies; another type of economy leverages geopolitical cooperation opportunities, adopting an embedded development strategy that combines foreign investment and differentiated competition, relying on external resources and technologies to facilitate industrial breakthroughs. This difference in strategic choices provides an important research perspective for observing the paths taken by late-developing economies to overcome development bottlenecks in global technological competition.

[0003] Against the backdrop of intensified global technological competition and shifting geopolitical landscape, methods for industrial digital management have gradually emerged and been implemented. By integrating geopolitical and industrial data, this approach provides comprehensive strategic decision-making support for countries and enterprises. Through this digital management framework, industrial policies can be optimized and evaluated in real time within a constantly changing external environment, ensuring the supply chain remains flexible and adaptable in the face of complex external conditions.

[0004] Existing industrial digital management methods, combined with advanced digital tools and data analytics, can dynamically capture and respond to changes in the global industrial environment. These methods utilize big data, artificial intelligence, and the Internet of Things (IoT) to perform cross-domain data fusion and intelligent analysis, providing governments, businesses, and policymakers with scientific and accurate decision-making support. For example, through digital twin technology and real-time monitoring systems, the operational status and potential risks of each link in the industrial chain can be assessed in real time; with the help of data mining and machine learning techniques, technological development trends can be predicted and analyzed, aiding in the formulation of forward-looking strategies. Therefore, industrial digital management methods not only enhance the ability to cope with complex environments but also help countries achieve technological catch-up and industrial upgrading, further driving the evolution of the global industrial landscape.

[0005] Based on the above technical process analysis, the existing processes mainly have several specific technical problems in the industrial digital management side: First, the underlying data modeling is mostly still the traditional time-region-indicator wide table or indicator warehouse structure. For objects such as policy tools, technology path nodes, and external events that have life cycles, stage changes, and many-to-many relationships, they are often simply recorded with a few fields (such as a type code and a status bit). It is rare to build separate entity tables that can be managed and associated separately. As a result, when you want to trace the evolution of a certain set of policy tools, the corresponding technology path, and the final performance results in a chain, you need to rely heavily on temporary associations and manual sorting, which limits the degree of automation.

[0006] Secondly, regarding time-dimensional processing, existing data flows typically only retain one occurrence time or statistical time. However, in reality, the same event often has multiple key time points simultaneously, such as release time, start execution time, technology implementation time, and effect evaluation time. The intervals and order between these time points are crucial to the analysis conclusions, but they are often not managed by the system in existing data integration and stream processing tasks. This leads to the need for repeated splicing using offline scripts when conducting subsequent strategy evolution reviews or process alignment. Thirdly, in the fusion of data at different time scales, environmental and constraint events are usually recorded daily or weekly, policy and investment indicators are statistically analyzed quarterly or annually, and process and equipment data are continuous streams at the second level or even finer. Existing systems often use simple summaries or alignment windows with fixed standards to piece together the data, which is weak in its ability to collaboratively express the hierarchical structure of long-cycle strategy - medium-cycle adjustment - short-cycle operation. This makes it difficult to quantify the process of phased advancement and step-by-step evolution of the technology path in detail.

[0007] Finally, at the analysis and decision support level, many platforms still rely mainly on static reports and conventional time series analysis. They lack the capabilities required for actual industry data management, such as multi-scenario simulation, comparison of different path branches, and strategy replay. They often can only be implemented temporarily in the form of one-off projects, making it difficult to develop into general data processing and analysis capabilities that can be reused repeatedly. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an industrial digital management method and system that integrates geopolitical environmental data, which can effectively solve the problems mentioned in the background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides an industrial digital management method integrating geopolitical environmental data, comprising: an industrial digital management platform reading targeted data from raw heterogeneous data and attaching it to pre-constructed basic management objects according to targeted time slices, and updating the basic management objects in pre-constructed TPE scenario units; the industrial digital management platform calculating geopolitical environmental volatility for geopolitical scenarios under targeted time slices, customizing an adaptive aggregation mode to aggregate targeted data, and evaluating the channel exposure vector and node structural feature parameters of each TPE scenario unit after aggregation; performing integrated geopolitical scenario simulation based on basic statistical features, geopolitical environmental volatility, channel exposure vector, and node structural feature parameters on the industrial digital management platform; filtering TPE scenario records in the TPE scenario units, inputting the filtered TPE scenario records into the DEA of the industrial digital management platform, the DEA decision unit outputting the corresponding industrial input matrix and industrial output matrix, and configuring the simulation prediction process of the geopolitical scenario in conjunction with the simulation simulation process to complete industrial digital management.

[0010] The second aspect of this invention provides an industrial digital management system integrating geopolitical environmental data, comprising: a TPE scenario unit construction module, used by the industrial digital management platform to read targeted data from raw heterogeneous data, attach it to pre-built basic management objects according to the targeted time slice, and update the basic management objects in the pre-built TPE scenario units; a TPE scenario unit evaluation module, used by the industrial digital management platform to calculate geopolitical environmental volatility for geopolitical scenarios under the targeted time slice, customize an adaptive aggregation mode to aggregate targeted data, and evaluate the channel exposure vector and node structure feature parameters of each TPE scenario unit after aggregation; an environmental scenario simulation and deduction module, used to perform integrated geopolitical scenario simulation and deduction on the industrial digital management platform based on basic statistical features, geopolitical environmental volatility, channel exposure vector, and node structure feature parameters; and a simulation prediction process configuration module, used to filter TPE scenario records in TPE scenario units, input the filtered TPE scenario records into the DEA of the industrial digital management platform, the DEA decision unit outputs the corresponding industrial input matrix and industrial output matrix, and configures the simulation prediction process of geopolitical scenarios in conjunction with the simulation and deduction process to complete industrial digital management.

[0011] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention provides an industrial digital management method and system that integrates geopolitical environmental data. The industrial digital management platform reads targeted data from the original heterogeneous data, attaches it to the pre-constructed basic management object according to the targeted time slice, and writes it into the corresponding TPE scene unit. Under the unified time dimension and management object system, a scene-based data base that can be indexed by region-path-environment-time is formed. Under each targeted time slice, the platform calculates the geopolitical environmental volatility for the geopolitical environmental scene, selects an adaptive time aggregation mode based on the geopolitical environmental volatility, performs time aggregation on the targeted data, and evaluates the channel exposure vector and node structure feature parameters of each TPE scene unit on the aggregation result. This simultaneously reflects the intensity of external environmental disturbances and the direction of pressure on the chain structure. The platform further uses basic statistical features, geopolitical environmental volatility, channel exposure vectors, and node structural feature parameters as simulation inputs to perform integrated geopolitical environmental scenario simulations on the target TPE scenario units. This is used to predict the change trajectory of key operational indicators under preset geopolitical environmental scenario combinations and time evolution constraints. Subsequently, effective data records are filtered at the TPE scenario unit level, and the filtered TPE scenario records are input into the DEA model to obtain the corresponding industry input matrix and industry output matrix. The simulation prediction process of geopolitical environmental scenarios is then configured in conjunction with the simulation input features, thereby completing a closed-loop industrial digital management process within the same framework, from historical efficiency assessment to resource and target constraint setting to scenario-based simulation prediction.

[0012] (2) By evaluating the channel exposure vector and node structural feature parameters of each TPE scenario unit, the present invention enables the platform to quantitatively characterize which channels the external geopolitical environment transmits to the internal industrial chain and which nodes on the chain are more structurally sensitive before efficiency analysis and simulation prediction. Thus, when the geopolitical environment scenario changes, it can identify high-exposure channels and key nodes in a targeted manner, support the configuration of differentiated control strategies for specific channels or specific nodes, and avoid the blind spots caused by relying solely on macro indicators for coarse-grained regulation.

[0013] (3) Throughout the process, the present invention reuses basic statistical features, geopolitical environmental fluctuations, channel exposure vectors and node structural feature parameters in multiple stages: the same set of features is used for both adaptive time aggregation and aggregation mode selection, as well as for the integrated setting of simulation process parameter correction and DEA output constraints, so that different functional modules share unified data expression and parameter meaning, thereby reducing the common problems of inconsistent parameter caliber and repeated modeling when multiple models are parallel, improving the reusability and maintainability of the overall computing link of the platform, and also reducing the expansion cost when adding new scenarios or new indicators.

[0014] (4) Compared with the industrial digital management technology that usually only evaluates efficiency and displays reports based on static operating data or a single KPI, this invention integrates the modeling and linkage configuration of geopolitical environment fluctuation, channel structure exposure, on-chain node structure characteristics, input-output efficiency analysis and scenario simulation process in a unified platform. On the one hand, it can explicitly introduce the influence of external environment and on-chain structure when evaluating efficiency, so that the efficiency results are closer to the real constraints. On the other hand, it can directly convert the DEA results into the upper limit of resource input and the output target range in the simulation process, and combine geopolitical environment scenario combination and time evolution constraints to make multi-scenario prediction. Thus, under the same method system, it can simultaneously support historical evaluation, structural diagnosis and forward-looking scenario induction, and improve the decision support capability of industrial digital management in complex geopolitical environments. Attached Figure Description

[0015] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0016] Figure 1 is a schematic flowchart of the method steps of the present invention; Figure 2 is a schematic diagram of the system module connection of the present invention; Figure 3 is a flowchart of the basic feature calculation of the present invention; Figure 4 is a flowchart of the DEA efficiency evaluation simulation prediction of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] Referring to FIG1, the first aspect of the present invention provides an industrial digital management method integrating geopolitical environmental data, comprising: an industrial digital management platform reading targeted data from raw heterogeneous data, attaching it to a pre-constructed basic management object according to the targeted time slice, and updating the basic management object in the pre-constructed TPE scene unit.

[0020] In this application, the aforementioned raw heterogeneous data refers to a collection of data generated by different business systems, different devices, or different external data sources that has not yet undergone unified cleaning and standardization. These data differ in their data sources, formats, and structures. Specifically, they include, but are not limited to: structured operation monitoring data and log data generated by production execution systems, equipment acquisition terminals, sensor gateways, etc.; configuration data, resource input data, and performance statistics data generated by enterprise management systems and project management systems; regional rule information, logistics and customs clearance information, supply chain event information, and macro-environmental indicators obtained from geographical environmental data sources, third-party data service platforms, or public data interfaces; and semi-structured or unstructured text data formed by policy texts, institutional documents, technical route descriptions, and report documents. Different raw heterogeneous data are inconsistent in their coding systems, time granularity, units of measurement, and field meanings. They need to be preprocessed and mapped by an industrial digital management platform before they can be linked to a unified basic management object system and participate in subsequent time aggregation, environmental analysis, and simulation management.

[0021] In this embodiment, the aforementioned targeted data refers to a set of target data pre-specified by the industrial digital management platform and used as the object of digital management and analysis. This data is used to focus on and continuously track specific management objects or operational processes under a unified data model. In this embodiment, the targeted data is preferably operational data from each link of the industry chain, including but not limited to production cycle time, capacity utilization rate, yield, work-in-process quantity, energy consumption level, delivery cycle, equipment status, and configuration parameters directly related to the aforementioned operational processes. This data is used for digital management and process control of the semiconductor industry chain's operational status.

[0022] It should be noted that in the industrial digital management platform of the present invention, the targeted data is not limited to the operational process data. Cost data, carbon emission data, quality inspection data, supply chain collaboration data, or other data types with management significance can also be designated as targeted data according to application needs. By linking different categories of target data to basic management objects and incorporating them into the same time and scenario dimensions for processing, unified digital management and analysis of multiple types of business objectives can be achieved.

[0023] Specifically, the basic management objects and TPE scenario units are pre-constructed as follows: a basic management object table is established in the data layer of the industrial digital management platform. The basic management objects include: regional management object table R, management tool object table T, technology path object table P, geo-environment scenario object table E, action channel object table C, and on-chain node object table N.

[0024] The aforementioned regional management object R table is used to store basic information at the regional or national level. Each region is assigned a unique R_ID and records the region name, the currency used, the price index version, the statistical caliber description, and metadata related to the statistical caliber, so as to provide a unified regional benchmark when comparing cross-regional data and conducting time series analysis.

[0025] The management tool object table T is used to store various resource configuration methods and management tools. Tool records are extracted from management rule data and resource configuration data, and a unique T_ID is generated for each tool record. The table also records attribute fields such as tool type, target action, coverage area, execution subject, and intensity parameters. At the same time, it is associated with the regional management object table R through the R_ID field to describe the differences in management tool configuration in different regions.

[0026] The technology path object P table is used to store industry technology routes and their phase divisions. Based on the technology route documents and process configuration data, the combination of path-phase is split into multiple technology path phase records. A unique P_ID is generated for each record, and information such as the path name, phase number, process level, corresponding industrial chain link, and associated node type are recorded. At the same time, it is associated with the corresponding region through the R_ID field to describe the layout of different regions on different technology path phases.

[0027] Geopolitical Environment Scene Object E Table: Used to store external environmental events and their categorized scene information. Geopolitical environmental events are grouped into several geopolitical environmental scenes according to preset rules. A unique E_ID is generated for each geopolitical environmental scene, and attribute fields such as scene type, affected area or affected node range, expected duration, direction of action and intensity level are recorded. These are used to calculate environmental volatility in subsequent steps and drive the adaptive adjustment of the time aggregation strategy.

[0028] The C table, representing the action channel object, stores information on various transmission channels through which the external environment affects the industrial chain. Based on the preset action channel definition, it generates a unique C_ID for each channel record and records the channel type and the set of quantitative indicators associated with that channel. The set of quantitative indicators includes at least the licensing index, supply concentration, weighted average cost of capital, tariff and non-tariff barrier levels, port accessibility, and customs clearance timeliness, which are used to quantitatively characterize the channel exposure vector of the TPE scenario unit in the subsequent process.

[0029] The on-chain node object N table stores node information for each link and process position in the industry chain. Based on the division of labor in the industry chain, each key link and process position is abstracted into a node. A unique N_ID is generated for each node, and attribute fields such as node type (including but not limited to EDA design, different process segments of wafer manufacturing, packaging and testing, system integration, etc.), the type of key equipment it depends on, whether it belongs to a high-licensing intensity node, and whether it has modular characteristics are recorded. At the same time, it is associated with the regional management object R table through the R_ID field to support the calculation of subsequent node structural characteristic parameters and regional difference analysis.

[0030] The industrial digital management platform is based on the regional management object table R, the management tool object table T, the technology path object table P, and the geo-environment scenario object table E. Under the time slice identifier time_slice, where the time slice represents the time segmentation rule defined in the platform in advance, it traverses each region according to the regional identifier R_ID, and selects the management tool identifier T_ID and geo-environment scenario identifier E_ID that are effective for the targeted region and its technology path stage within the time slice. Combined with the technology path stage identifier P_ID and its associated on-chain node set, a corresponding TPE scenario unit record is generated for each (R_ID, T_ID, P_ID, E_ID, time_slice) combination. The scenario unit reserves the target indicator field, the channel exposure vector field, and the node structure feature field.

[0031] Example 1: Taking the expansion of mature process technology in China under a strong blockade as an example: ① Establish six types of basic objects: Regional object R: R_ID=CN, Name=China; Management tool object T: T_ID=T_CN_FUND, Name=National Integrated Circuit Industry Investment Fund, Function description=Provides equity investment and financial support to wafer manufacturing companies and key equipment / material companies; Technology path stage object P: P_ID=P_CN_MATURE_28, Name= The 28nm and above mature process capacity expansion stage describes the rapid expansion of capacity, improved yield, and economies of scale on mature processes; the geopolitical environment scenario object E: E_ID=E_BLOCKADE_HIGH, name=high-intensity external technology control scenario, description=key equipment and advanced processes are restricted, and the whole is under blockade pressure; the on-chain node object N: (only one main node is given here) N_ID=N_WAFER_MFG_MATURE, name=mature process wafer manufacturing node, attributes=manufacturing stage, medium licensing intensity, highly sensitive to equipment and material channels; the time slice object time_slice: Time_ID=Y2019, year=2019 (can also be replaced with a quarter, here we use the whole year as an example);

[0032] ② Use these six types of objects to construct a TPE scenario unit: TPE_1: Region China (CN), Management tool = National Integrated Circuit Industry Investment Fund (T_CN_FUND), Technology path stage = Mature process 28nm capacity expansion (P_CN_MATURE_28), Geopolitical environment scenario = High-intensity technology control (E_BLOCKADE_HIGH), Time slice = 2019 (Y2019), Nodes involved = Mature process wafer manufacturing nodes (N_WAFER_MFG_MATURE).

[0033] In the database, it's a TPE record, roughly looking like this (logically):

[0034] This is used to characterize the digital management scenario in which China, under the circumstances of high-intensity external technology control, relied on the National Integrated Circuit Industry Investment Fund to promote the expansion of production capacity and yield improvement of mature processes of 28nm and above in 2019. Under this TPE scenario unit, subsequent operational data such as mature process capacity, capacity utilization rate, yield performance, equipment utilization rate, and related geopolitical environment data are attached to support subsequent environmental fluctuation calculation, channel exposure assessment, node structure feature extraction, and simulation prediction.

[0035] Example 2: Taking India's semiconductor industry as an example, R_ID=IN (India); T_ID=T_IN_SEMICON (Semicon India plan); P_ID=P_IN_ATMP_ENTRY (packaging and testing / ATMP entry stage); E_ID=E_FRIEND_SHORING (outsourcing and investment promotion scenario); N_ID=N_ATMP (packaging and testing node); Time_ID=Y2022 (2022).

[0036] Obtained an Indian TPE:

[0037] This is used to characterize India's digital management scenario in the context of friendly outsourcing, relying on Semicon India's plan to enter the semiconductor industry chain through packaging and testing / ATMP in 2022. Under this TPE scenario unit, subsequent operational data such as packaging and testing capacity, number of projects, yield performance, and related geopolitical environmental data are attached to support subsequent environmental volatility calculation, channel exposure assessment, node structure feature extraction, and simulation prediction.

[0038] It should be explained that the pre-construction process of the above-mentioned TPE scenario units also includes: the industrial digital management platform establishing a tiered time-series dimension table in the data layer for unified time management of various business data and TPE scenario units: among which, a year dimension table Time_Year is constructed to record the year and its attribute information; a period dimension table Time_Period is constructed to record the quarter or month and is linked to the Time_Year through foreign keys; and a day / hour dimension table Time_Day and / or Time_Hour is constructed to record the specific day or hour and is linked to the Time_Period and Time_Year.

[0039] Building upon this foundation, a time-dimensional foreign key field is added to the TPE scene unit records. This associates the time slice identifier (time_slice) of each TPE scene unit record with the corresponding Time_Period record, and further associates Time_Period with the corresponding Time_Year and its covered day / hour range. This achieves unified positioning of TPE scene unit records across different time levels. The platform further constructs a multi-dimensional data index based on Time_Year, Time_Period, and TPE scene unit records, enabling rapid querying and aggregation of TPE scene units and their associated data using a combination of dimensions (Year, Period, Region_ID, T_ID, P_ID, E_ID). This provides unified time support for subsequent cross-year and cross-quarter integrated geopolitical environment analysis and simulation management.

[0040] The industrial digital management platform reads targeted setting data with timestamps from the original heterogeneous data, performs time standardization processing on the targeted setting data, generates hierarchical time dimensions of the targeted setting data, and links the targeted setting data to the basic management objects according to the hierarchical time dimensions, so that the targeted setting data are associated with the corresponding regional management objects, management tool objects, technology path objects, and geographical environment scenario objects respectively.

[0041] In this invention, the industrial digital management platform first obtains targeted data with timestamps from different business systems, devices, or external data sources. Because the time format, time zone, and granularity of these data are different, all times are first converted into the same standard time according to the platform's unified time standard. Incidentally, the data is hierarchically divided by year, quarter / month, day, and hour to build a time tree. Then, according to unified coding rules, each target data with standardized time is attached to this time tree and the pre-built basic management objects. That is to say, a piece of data will know which year / quarter it occurred in, which region it belongs to, which management tool it uses, which stage of the technology path it is in, and what geographical environment it was generated in. In this way, it can be uniformly queried, statistically analyzed, and simulated by time slice, region, tool, path, and environmental scenario.

[0042] The industrial digital management platform updates the pre-built TPE scene unit records by combining regional identifiers, management tool identifiers, technology path stage identifiers, geographical environment scene identifiers, and time slice identifiers, so that each TPE scene unit record carries targeted setting data that matches the management object and time dimension under the targeted time slice.

[0043] For each combination of [regional identifier, management tool identifier, technology path stage identifier, geopolitical environment scenario identifier, and time slice identifier], the industrial digital management platform retrieves targeted setting data records from the standardized targeted setting dataset whose timestamps fall within the time slice and match all the aforementioned identifier conditions. Based on preset field mapping rules, the platform writes or associates the targeted setting data with the corresponding TPE scenario unit's operation indicator fields, configuration parameter fields, and / or tag fields. If no valid targeted setting data is found for a particular combination within the target time slice, the relevant fields in the corresponding TPE scenario unit are marked as null or missing. Through this update process, each TPE scenario unit record carries targeted setting data matching its region, management tool, technology path stage, and geopolitical environment scenario under its corresponding targeted time slice, providing a one-to-one and time-consistent data foundation for subsequent environmental fluctuation calculations, basic statistical feature extraction, and integrated geopolitical environment scenario simulation management.

[0044] It should be explained that the above field mapping rules are used to map targeted data fields from different business systems to the unified field system of the industrial digital management platform. Specifically, this includes, but is not limited to, one or more of the following types of rules: field name mapping rules, used to convert field names in the source system to platform standard field names, for example, mapping the "LotOutput" and "GoodDies" fields in the production execution system to the platform standard fields "number of wafers produced" and "number of qualified chips", respectively; unit conversion rules, used to unify the measurement units of the source fields to the platform's preset units, for example, converting the energy consumption field from "kWh / day" to "kWh / wafer", and the capacity field from "wafer / month" to "wafer / time slice"; time field mapping rules, used to map the timestamp field in the source data to the platform's time slice identifier field, and other similar rules.

[0045] In this application, a targeted time slice refers to a time slice selected by the current data processing step as the object for data attachment and scene update within a pre-divided set of time slices. Specifically, the industrial digital management platform first divides the entire cycle operation time into several preset time slices (e.g., by year, quarter, or month) based on a unified time benchmark. When performing TPE scene unit update operations, each time slice in the preset time slice set is selected as a targeted time slice in turn. Under the current targeted time slice, the retrieval, filtering, and attachment processing of the targeted data are completed. After processing is completed, the next preset time slice is set as the new targeted time slice to continue execution, thereby ensuring that all preset time slices are processed one by one as targeted time slices under the same rules, achieving complete coverage and orderly management of TPE scene units in the time dimension.

[0046] The industrial digital management platform calculates geopolitical environmental volatility for geopolitical scenarios under targeted time slices, customizes an adaptive aggregation mode to aggregate targeted data, and evaluates the channel exposure vector and node structure feature parameters of each TPE scenario unit after aggregation.

[0047] Specifically, geopolitical environment volatility is calculated for geopolitical environment scenarios under targeted time slices, and a customized adaptive aggregation mode is used to aggregate targeted data. The specific analysis process is as follows: geopolitical environment data sequences under each targeted time slice are read from the original heterogeneous data. Under each targeted time slice, the industrial digital management platform calculates the corresponding geopolitical environment volatility based on the geopolitical environment data sequences in the current targeted time slice and adjacent targeted time slices.

[0048] The geopolitical environmental data includes at least records of technology control events, adjustments to tariffs and non-tariff measures, changes in the supply of key equipment and materials, port and customs clearance timeliness data, logistics accessibility indices, and environmental indicators related to capital costs and market access. The platform then standardizes these environmental indicators, normalizing indicators with different dimensions into a dimensionless standard indicator sequence.

[0049] For each targeted time slice s, within a time window W(s) that includes the current targeted time slice and several adjacent targeted time slices, various geopolitical environmental indicators x are evaluated. j (t) is standardized, where there are M types of environmental indicators in the geopolitical environment (such as the intensity of technology control, tariff level, equipment delivery cycle, etc.), and the original value of the j-th type of environmental indicator at time t is denoted as: x j (t), , t∈W(s), based on the standardized index sequence x' j (t) Calculate the average deviation D j (s): Fluctuation amplitude F j (s): Mutation intensity J j (s): , where t - The time slice preceding t is represented as t.

[0050] And according to the preset weight α in the industrial digital management platform j ,β j ,γ j The combination yields the single-indicator volatility V. j (s), specifically: V j (s)=D j (s)*α j +F j (s)*β j +J j (s)*γ j Then, according to the importance weight λ of the indicators j For each V j (s) Weighted summation and normalization yield the comprehensive geopolitical environmental volatility Venv(s) corresponding to the targeted time slice s, specifically: ,in It is expressed as a normalized function, thus using a unified scalar to characterize the combined strength of the internal and external geopolitical disturbances of the time slice and its adjacent time slices.

[0051] The geopolitical environment volatility of the geopolitical environment scenario is compared with a predefined geopolitical environment volatility threshold. If the geopolitical environment volatility of the geopolitical environment scenario is less than or equal to the geopolitical environment volatility threshold, the aggregation mode is set to the first aggregation mode; otherwise, the aggregation mode is set to the second aggregation mode.

[0052] The aforementioned geopolitical environment volatility refers to the environmental disturbance intensity parameter obtained by the industrial digital management platform after comprehensively quantifying the level deviation, fluctuation range, and abrupt change intensity of various environmental indicators such as technology control intensity, tariff and non-tariff measures, key equipment and material supply, port and customs clearance timeliness, logistics accessibility, capital costs, and market access based on the geopolitical environment data sequence within the targeted time slice and its adjacent time slices. It is used to characterize the overall instability of the external geopolitical environment within a given time window with a single scalar.

[0053] Based on the comparison between the geopolitical environment volatility and the preset volatility threshold, the platform adaptively selects the aggregation mode and binds time aggregation windows of different lengths and granularities to different aggregation modes. By using the geopolitical environment volatility as a control parameter to limit the size of the time aggregation window, the industrial digital management platform can automatically expand the time scale to improve statistical stability in stable environmental ranges and automatically shorten the time scale to retain impact details in areas of severe environmental fluctuations, thereby achieving linkage and matching between the time aggregation scale and the intensity of geopolitical environmental disturbances.

[0054] Based on the time aggregation window corresponding to the aggregation mode, the target setting data within the window is aggregated over time, and the basic statistical features of the time-aggregated target setting data are calculated.

[0055] The time aggregation window corresponding to the aggregation mode refers to the time range and granularity pre-configured by the industrial digital management platform for the aggregation time according to different aggregation modes: the platform uses time slices as the smallest time unit (e.g., "quarters" as time slices) and pre-sets a long-cycle window length threshold N. long and short period window length threshold N short And satisfy N long >N short ≥1. When the aggregation mode is set to the first aggregation mode, the number N of the targeted time slices covered by the time aggregation window. win Satisfying N win ≥N long For example, when using quarters as time slices, N can be... long Setting it to 4 means the time aggregation window covers four or more consecutive quarters, which is equivalent to summarizing and processing targeted setting data and operational data within the same relatively stable environmental period on an annual or multi-year basis. When the aggregation mode is set to the second aggregation mode, the number N of the targeted time slices covered by the time aggregation window... win Satisfying 1≤Nwin ≤N short For example, when using quarters as time slices, N can be... short Setting it to 1 or 2 means the time aggregation window only covers the current quarter or the current quarter plus one adjacent quarter. If months are used as time slices, short-cycle windows can be configured to 1-3 months, with a significantly shorter time span than long-cycle windows. By explicitly binding two configurable parameters—"number of covered time slices" and "corresponding time length"—to different aggregation modes, the size of the time aggregation window can be controlled by the degree of geopolitical environmental volatility. This results in a longer window for a more stable geopolitical environment and a shorter window for a more volatile one, making the adaptive linkage between the time aggregation scale and the intensity of environmental disturbances clearer and more quantifiable.

[0056] The specific calculation process of the basic statistical features of the aforementioned targeted data is shown in Figure 3. Figure 3 is a flowchart of the basic feature calculation of the present invention, which describes how the industrial digital management platform first reads the original heterogeneous data and performs time standardization and time slice division; then it determines whether targeted data exists in the current targeted time slice. If not, the time slice is marked as missing data and the processing of the current time slice ends; if it exists, the targeted data is attached to the pre-constructed basic management object according to the targeted time slice, the corresponding TPE scene unit is updated, and the geopolitical environment data sequence under the scene is extracted to calculate the geopolitical environment volatility V_env. The platform selects between long window aggregation and short window aggregation modes based on whether V_env is less than or equal to a preset threshold. Within the selected time aggregation window, the targeted data is aggregated, and the basic statistical features are calculated. Based on this, the channel exposure vector and node structure feature parameters are calculated, and finally the calculation of the basic features under the current targeted time slice is completed.

[0057] The industrial digital management platform targets the regional identifier, management tool identifier, technology path stage identifier, geographical environment scenario identifier, and time slice identifier carried by each TPE scenario unit. It then filters out all data records whose timestamps fall within the time aggregation window from the targeted data and groups the filtered data records according to the targeted data categories.

[0058] Based on the grouped data records, basic statistical features are calculated for various target indicators of each TPE scene unit within the time aggregation window. The basic statistical features include one or more of the following: mean, standard deviation, maximum value, minimum value, linear trend representation, and number of anomalies of various target indicators of the TPE scene unit. The basic statistical features are then written into the target indicator field of the corresponding TPE scene unit.

[0059] Among them, various target indicators refer to the set of target indicators selected as the statistical and analytical objects for each TPE scenario unit within the time aggregation window. In this embodiment, they are preferably operational indicators, including but not limited to capacity utilization rate, yield rate, work-in-process quantity, equipment utilization rate, unit output energy consumption, and order on-time delivery rate. At this time, the mean, standard deviation, maximum value, minimum value, linear trend representation, and number of anomalies in the basic statistical characteristics are used to characterize the average operating level, fluctuation degree, extreme performance, change direction, and abnormal fluctuation frequency of the above operational indicators within the time aggregation window.

[0060] In another exemplary implementation, the target indicator can also be a cost-related or performance-related indicator, such as unit output manufacturing cost, unit output carbon emissions, project investment return rate, etc. In this case, the basic statistical features are used to characterize the cost stability, cost peak risk, or performance fluctuation of these cost / performance indicators within the window.

[0061] In another exemplary embodiment, the target indicators can also be supply chain and risk indicators, such as the delivery cycle of key materials, the number of supply interruptions, and the dependence on a single supplier. In this case, the basic statistical features are used to characterize the average level, volatility, extreme delays, and frequency of abnormal interruptions of supply assurance capabilities over time. This allows the present invention to perform digital management of operational indicators and extend analysis of other types of indicators such as cost, environment, and supply risks within the same time aggregation and statistical framework.

[0062] Through the aforementioned time aggregation process, on the one hand, high-frequency targeted data spanning days, weeks, or months can be reduced to a small number of aggregated results on the time scale corresponding to the current environmental intensity, thereby reducing subsequent computational complexity and noise sensitivity; on the other hand, it provides a unified window-level input for basic statistical feature calculation, enabling the platform to calculate basic statistical features such as time average, volatility, extreme values, and number of anomalies for each operating indicator on the aggregated time scale, thus characterizing the operating status of TPE scene units at a time granularity that is neither too fragmented nor too smooth.

[0063] Specifically, the channel exposure vector and node structure feature parameters of each TPE scenario unit are evaluated. The specific analysis process is as follows: Based on the preset mapping relationship between the geo-environment scenario object and the function channel object table, the industrial digital management platform calculates the structural weight of each function channel under the TPE scenario unit by combining the association relationship between the on-chain node object table and the technology path object table.

[0064] The mapping relationship refers to the fact that the industrial digital management platform pre-configures the impact distribution of each geopolitical scenario on various action channels, and establishes a one-to-one correspondence between each geopolitical scenario object E_ID and the channel identifiers C_ID in the seven action channels (technology availability, equipment and material supply, capital and finance, market access, talent and education, standards and intellectual property, and logistics and geography). This is used to characterize the theoretical strength of the transmission of external environmental changes to the industrial chain through each action channel in the geopolitical scenario.

[0065] The relationship refers to the following: The platform records the set of on-chain nodes covered by each technology path stage P_ID in the technology path object table, and records the industry segment, licensing strength attribute, modularity degree and sensitivity parameters of each action channel for each node N_ID in the on-chain node object table. By referencing the node identifiers in the on-chain node object table in the technology path object table, a structured description of which nodes constitute a certain technology path stage and how sensitive each node is to different action channels is realized.

[0066] Let the current TPE scenario unit be denoted as u, which corresponds to: geopolitical environment scenario E(u), technology path stage P(u), node set N(u) (obtained from path-node association), and channel set C={c1,...,c K}

[0067] Preset a scene-channel mapping matrix: M E，C =(m e，c ), where: m e，c ≥0 indicates the scene basis weight of the geopolitical environment scene e to channel c. And for the current TPE scene unit u, the scene basis weight is b. c (u)=m E(u)， c。

[0068] To characterize the sensitivity of nodes on different chains to different channels, a channel-node sensitivity function is preset: k c，n =f c (attr(n)), where: n∈N(u) is a node under path stage P(u); attr(n) is the attribute vector of node n (e.g., whether it is a high-permission intensity node, a high-concentration node, or highly dependent on logistics channels, etc.); fc To calculate the sensitivity of channel c, output a non-negative value k. c，n , represents the structural sensitivity of node n under channel c.

[0069] For the current node set N(u) of TPE, first aggregate to obtain the node structure contribution of channel c: By combining the scene basis weights with the node structure contribution, the original structural strength of channel c under the TPE scene unit u is obtained: .

[0070] To facilitate comparison of the relative importance of each channel within this TPE scene unit, the original strengths of all channels are normalized to obtain the structural weights w. c (u) represents the structural weight of channel c in the TPE scenario unit u, indicating the relative importance of channel c in the overall channel structure under the given geopolitical environment scenario and the node structure of this technology path stage.

[0071] Based on the basic statistical characteristics, environmental response feature parameters are constructed. Through the mapping relationship between the preset environmental response feature parameters and dynamic feature coefficients in the industrial digital management platform, dynamic feature coefficients are obtained by matching the environmental response feature parameters. The industrial digital management platform uses the product of the structural weight of each action channel and the corresponding dynamic adjustment coefficient as the channel exposure vector of that action channel in the TPE scene unit, and writes it into the channel exposure vector field of the TPE scene unit.

[0072] The industrial digital management platform extracts the structural information of each node from the on-chain node object table based on the set of on-chain nodes covered by the TPE scenario unit, counts structural feature indicators, combines the structural feature indicators to form node structural feature parameters, and writes them into the node structural feature field of the TPE scenario unit.

[0073] The industrial digital management platform first reads the set of on-chain nodes pre-attached in the TPE scenario unit, uses the node identifier of each node in the set as the search condition, and extracts the structural information of each node from the on-chain node object table one by one. The structural information includes at least the process link to which the node belongs (such as design, wafer manufacturing, packaging and testing, etc.), its hierarchical position in the industry chain (upstream / midstream / downstream), licensing strength attribute (high licensing strength node or general node), modularity attribute (substitutable / difficult to substitut), supply role attribute (single key supply node or multi-source node), and node importance index (such as key node mark or node concentration score).

[0074] After extracting node information, the platform uses the set of nodes covered by the TPE scenario unit as the statistical object, counts the number of nodes of different categories, and calculates the corresponding proportions based on the counting results. For example, it calculates the proportion of high-permission intensity nodes and modular nodes by using the total number of nodes as the denominator, calculates the proportion of nodes at each level by dividing the number of upstream, midstream, and downstream nodes by the total number of nodes, calculates the proportion of key nodes by dividing the number of nodes marked as key nodes by the total number of nodes, and calculates the proportion of single key supply nodes by dividing the number of nodes marked as single key supply nodes by the total number of nodes, thus obtaining a set of specific structural characteristic indicators.

[0075] Subsequently, the platform sequentially writes the aforementioned structural feature indicators into the corresponding field positions in the node structural feature record according to the predefined field order. For example, the proportion of high-permission intensity nodes is written into field F1, the proportion of modular nodes into field F2, the proportions of upstream / midstream / downstream nodes into fields F3, F4, and F5 respectively, and the proportions of key nodes and single key supply nodes into fields F6 and F7 respectively, thus forming a node structural feature parameter vector with a fixed field order. Finally, the platform associates or writes the node structural feature parameter vector as a whole into the target TPE scene unit, which is used as an input feature to characterize the structural characteristics of the TPE scene unit chain in geopolitical environment scene simulation.

[0076] Based on basic statistical characteristics, geopolitical environment volatility, channel exposure vectors, and node structural characteristic parameters, the integrated geopolitical environment scenario simulation is conducted on the industrial digital management platform.

[0077] The simulation management layer of the industrial digital management platform receives the target TPE scene unit selected by the user. Based on the target TPE scene unit, the simulation management layer loads the corresponding geopolitical environment scene simulation process template and uses the basic statistical features, geopolitical environment volatility, channel exposure vector and node structure feature parameters already written in the target TPE scene unit as simulation input features to dynamically correct the geopolitical environment scene simulation process template.

[0078] After selecting a target TPE scenario unit and loading the corresponding geopolitical environment scenario simulation process template, the industrial digital management platform first uses the pre-written basic statistical characteristics in the TPE scenario unit as the operating baseline to calibrate the initial values ​​of each operating indicator in the simulation process template: for the k-th type of operating indicator in the simulation process template, its initial value is directly set to the mean μ of the operating indicator of the TPE scenario unit within the time aggregation window. k Its allowable fluctuation range is set to [μ k -λ k σ k , μ k +λ k σk ], where σ k Let λ be the standard deviation of this performance indicator within the time aggregation window. k This is the preset fluctuation coefficient.

[0079] Subsequently, using the geopolitical environmental volatility Venv calculated by the TPE scene unit as the environmental disturbance intensity parameter, the preset environmental disturbance amplitude A0, disturbance occurrence frequency f0, and simulation time step Δt0 in the simulation process template are scaled proportionally. Specifically, the environmental disturbance amplitude is adjusted to A = A0 * (1 + *Venv), adjust the perturbation frequency to f=f0*(1+ *Venv), adjust the simulation time step to Δt=Δt0 / (1+ *Venv), where This is the preset non-negative scaling factor.

[0080] Furthermore, the channel exposure vector e={e} of this TPE scene unit c As the channel allocation weight, it represents the proportion of the original disturbance injected into each active channel in the simulation process template. and conductivity coefficient To perform a reallocation, specifically: first press p c '= *(1+δ(e c - )), k c '= *(1+η(e c - The correction values ​​for each channel are obtained, where... The average exposure of each channel is δ, and η are preset adjustment coefficients. Then, {p} c The values ​​are normalized to a sum of 1, which is used as the actual perturbation injection ratio in the simulation process.

[0081] Meanwhile, the proportion of high-permission strength nodes r in the node structure characteristic parameters of this TPE scene unit lic Modular nodes account for a certain percentage (r) mod Key node percentage r key The parameters are the upper limit of the capacity of each node in the simulation template, adjusted according to an explicit linear correction rule. Buffer coefficient and fault sensitivity Make adjustments. , and This indicates the initial capacity limit, initial buffer coefficient, and initial fault sensitivity of node c in the simulation template: the capacity limit is set to Cap. n = *(1- *r key Set the buffer factor to Buf. n = *(1+θ2*r mod Set the fault sensitivity to Sens n = *(1+θ 3* r lic ), where θ1, θ2, and θ3 are preset node-level adjustment coefficients, and the backup path switching rule is triggered when the node is a critical node.

[0082] Through the above process of quantitatively calibrating the initial state of operation with mean and standard deviation, scaling the disturbance parameters proportionally with the geopolitical environment fluctuation, linearly correcting the channel injection ratio and conduction coefficient with the channel exposure vector, and explicitly correcting the capacity and sensitivity with the node structure characteristic parameters, the instantiated simulation process has clear adjustment and correction amounts at the parameter level, adaptively fitting the historical operating level, environmental intensity, channel pressure structure and on-chain node configuration of the target TPE scene unit.

[0083] The revised geopolitical environment scenario simulation process template simulates and extrapolates the trajectory of target index changes for targeted TPE scenario units through preset geopolitical environment scenario combinations and time evolution constraints.

[0084] The aforementioned geopolitical scenario combinations and temporal evolution constraints refer to the industrial digital management platform organizing various geopolitical scenarios of different intensities and types into one or more scenario sequences based on a predefined geopolitical scenario object table. Explicit constraints are then placed on the order of appearance, duration, and switching conditions of each scenario on the timeline. For example, the platform can configure scenario sequences consisting of normal scenarios, moderate disturbance scenarios, high-intensity disturbance scenarios, and recovery scenarios. It can specify the start and end time slices or duration steps for each scenario segment and stipulate that a scenario will switch to another when a certain environmental indicator threshold is reached. Simultaneously, it can set temporal evolution constraint parameters such as maximum cumulative disturbance intensity and minimum recovery period for different scenario combinations. These parameters limit the evolution path and speed of the geopolitical environment from a stable phase to an impact phase and then back to a relatively stable phase. Thus, during simulation, discrete scenario sequences and time series rules are jointly used to characterize the state trajectory of the external geopolitical environment over time.

[0085] Under the revised geopolitical environment scenario simulation template and the pre-configured geopolitical environment scenario combinations and time evolution constraints, the industrial digital management platform takes the target indicator state of the target TPE scenario unit at the initial moment as the simulation starting point. It sequentially traverses each environmental segment in the scenario sequence according to the simulation time step. Within each time step, it iteratively updates the value of the target indicator based on the current geopolitical scenario, channel parameters, and node structure parameters, and records the discrete sequence of target indicator changes over time. Specifically, in each simulation time step, the platform reads the environmental disturbance intensity and channel transmission parameters corresponding to the current scenario, calls the simulation equation to calculate the new values ​​of capacity utilization, yield, delivery cycle, cost, or risk-related target indicators after that time step, and uses these values ​​as the input state for the next time step. This process is repeated until the entire scenario sequence is traversed or a preset termination condition is met. Through this time-step numerical iteration process, the platform obtains the target indicator change trajectory of the target TPE scenario unit under different geopolitical environment scenario combinations and time evolution constraints, which is used for subsequent performance evaluation and decision-making scheme comparison.

[0086] TPE scenario records are filtered in the TPE scenario unit. The filtered TPE scenario records are input into the DEA of the industrial digital management platform. The DEA decision unit outputs the corresponding industrial input matrix and industrial output matrix. Combined with the simulation and deduction process, the simulation prediction process of the geopolitical environment scenario is configured to complete the industrial digital management.

[0087] Specifically, the DEA decision unit outputs the corresponding industry input matrix and industry output matrix, and combines the simulation prediction process of the geopolitical environment scenario with the simulation deduction process. The specific analysis process is as follows: The simulation management layer of the industrial digital management platform receives the filtering conditions set by the user, including preset regional conditions, time segment conditions and geopolitical environment scenario constraints. It filters out several TPE scenario records from each TPE scenario record and records them as a sample set. The sample set is used as the input of the DEA decision unit and outputs the industry input matrix and industry output matrix of the sample set.

[0088] The industrial input matrix and industrial output matrix are combined with the simulation input features and configured in an integrated manner. Based on the combination of geopolitical environmental scenarios and time evolution constraints, the upper limit of industrial resource input, the output target range, the intensity of geopolitical environmental disturbances and the transmission coefficients of each action channel are set in an integrated manner to complete the simulation prediction configuration process of geopolitical environmental scenarios.

[0089] After selecting a target TPE scenario unit, the industrial digital management platform first reads the decision unit row of the TPE scenario unit from the industrial input matrix and industrial output matrix output by the DEA decision unit. It then combines the values ​​of each input element in the input matrix with the basic statistical features in the simulation input features to calculate the benchmark input level and its upper and lower limits for the corresponding resource dimension, thereby setting the upper and lower limits of industrial resource input. At the same time, it combines the values ​​of each output indicator in the output matrix with the target operation indicator features in the simulation input features to determine the target output benchmark value and allowable deviation range of each target indicator during the simulation process, forming the output target range.

[0090] Input vector x(u)=(x i (u)), where i=1.. , The number of input factors (e.g., how many dimensions are involved in capital investment, equipment capacity, manpower, energy consumption, etc.). Output vector y(u) = ( (u)), where =1.. ,for The number of output indicators (e.g., output value, yield, export volume, added value, etc., how many dimensions are there). Resource and target constraints are set accordingly. ; ;in , This is a preset fluctuation coefficient.

[0091] Building upon this foundation, the platform further integrates the geopolitical environment scenario combination and temporal evolution constraints corresponding to the current TPE scenario unit. It uses geopolitical environment volatility parameters to correct the intensity of environmental disturbances during the simulation process, and channel exposure vectors to correct the transmission coefficients and disturbance distribution ratios of each active channel. This allows for the simultaneous setting of resource input boundaries, output target ranges, environmental disturbance intensity, and channel transmission coefficients within the same configuration step. The platform then uniformly writes the resource constraint parameters, target constraint parameters, environmental disturbance parameters, and channel transmission parameters obtained from the above integrated settings into the simulation configuration object corresponding to the TPE scenario unit. It also determines the simulation start and end time slices and scenario switching rules according to the geopolitical environment scenario combination and temporal evolution constraints. This forms a geopolitical environment scenario simulation prediction configuration that can be directly called by the simulation engine, providing a complete parameter basis and temporal logic constraints for subsequent automated simulation prediction of target indicator changes in the target TPE scenario unit under different geopolitical environment sequences.

[0092] The simulation management layer of the industrial digital management platform uses each input element column in the industrial input matrix as the initial parameters for the industrial resource allocation boundary and input intensity in the simulation prediction process. Specifically, the value range of each input element column is limited to the upper and lower limits of the allocation of various resource dimensions such as funds, production capacity, equipment capacity, human resources and materials in the simulation process, and its actual value in the target decision unit is used as the resource input level at the initial moment of the simulation.

[0093] Each output index in the industry output matrix is ​​listed as the target output parameter and performance evaluation threshold in the simulation prediction process.

[0094] Specifically, the output indicators reflected in each output indicator column, such as output value, yield, delivery capacity, and energy efficiency level, are used as target ranges or benchmark levels to be reached or compared during the simulation process. After the simulation is completed, the industrial digital management platform compares the simulation output results with the output indicator column (or its target range) given by the DEA model to determine whether each simulation scenario meets the feasibility constraints under the given input constraints and DEA target output requirements, and sorts and filters the different simulation scenarios accordingly.

[0095] It should be explained that the industrial digital management platform constructs the input and output matrices required for the DEA model, forming a DEA dataset with each TPE scenario unit as a row and input / output indicators as columns. The platform uses an input-oriented BCC-DEA model to evaluate the efficiency of the decision-making unit, and the specific process is shown in Figure 4. Figure 4 is a flowchart of the DEA efficiency evaluation simulation prediction of this invention.

[0096] The diagram first selects the target TPE scenario unit set as the object to be analyzed, and verifies the data integrity of the selected scenario units. If the data does not meet the requirements of DEA and simulation, a data missing prompt is output and the process ends. If the data is complete, a DEA input matrix and output matrix are constructed based on the set, and the input-oriented BCC-DEA model is run to obtain the efficiency score of each scenario unit. It is then determined whether the efficiency score is less than 1 or whether there is a non-zero slack. If so, the savings rate of each input indicator and the gap rate of each output indicator are calculated. Otherwise, the baseline input / output value is directly adopted, thereby forming an industry input matrix and industry output matrix containing efficiency correction information.

[0097] Subsequently, the platform reads the geopolitical environment volatility V_env, channel exposure vector, and node structure characteristic parameters bound to the target TPE scenario unit. In the simulation process template, it sets the upper and lower limits of resource input, the output target range, and simulation parameters such as environmental disturbances and channel transmission in an integrated manner. After the parameter settings are completed, the platform runs the simulation prediction according to the preset geopolitical environment scenario combination and time evolution constraints, obtains the simulation results of key operation indicators, and determines whether each simulation indicator falls within the target range: if it exceeds the target range, it outputs risk warnings and adjustment suggestions; if it does not exceed the target range, it outputs normal operation evaluation results, thereby completing the DEA efficiency evaluation and scenario-based simulation prediction for the TPE scenario unit.

[0098] It should be explained that the DEA efficiency assessment function is a derivative implementation of the industrial digital management platform of this invention, built upon its basic data integration and feature calculation capabilities. Based on the platform's completion of standardized processing of raw heterogeneous data, TPE scenario unit construction, and targeted indicator aggregation, the input and output indicators obtained from the aggregation of each TPE scenario unit are further organized into the input and output matrices required for DEA. Then, the input-oriented BCC-DEA model is invoked to evaluate the relative efficiency of the decision-making units corresponding to each TPE scenario unit, outputting efficiency scores and diagnostic information such as input redundancy and output gaps. Therefore, DEA efficiency assessment is not a necessary basic module of the platform, but rather a dedicated implementation for input-output efficiency analysis, extended from the platform's existing data foundation and feature system.

[0099] In an exemplary embodiment, the industrial digital management platform uses the semiconductor industries of China and India as comparison objects. It defines a combination of regional identifier and time slice identifier (e.g., "China-2015...2024" "India-2015...2024") as a DEA decision unit. Each decision unit is obtained by aggregating relevant TPE scenario units within that year according to preset rules. Input and output indicators are extracted from the aggregated TPE data and written into the DEA input and output matrices, respectively. The input indicators for the Chinese side include capital investment, wafer fab CapEx, R&D expenditure, equipment and material investment, labor and energy consumption levels, etc. The output indicators include wafer capacity (converted to 12-inch equivalent), mature process output, industry output value, export scale, and technology-related indicators. For India, the platform constructs input indicators such as the number of effective patents, design support funds, infrastructure expenditures, labor and energy consumption levels, and output indicators such as packaging and testing output value, design service revenue, number of foreign-invested projects and total industrial output value. After standardization and dimensional conversion of the above indicators, the platform calls the input-oriented BCC-DEA model to solve the efficiency score for each "region-year". Based on the optimal solution, it calculates the saving space of each input element and the improvement space of each output indicator. The efficiency score, input saving rate and output gap rate are written back to the China-India TPE scenario unit of the corresponding year. This not only forms the input-output efficiency time trajectory of the industrial paths of the two countries, but also provides the constraint input of the upper limit of resource input and the range of output target for subsequent geopolitical environment scenario simulation configuration.

[0100] Referring to FIG2, the second aspect of the present invention provides an industrial digital management system integrating geopolitical environmental data, including: a TPE scene unit construction module, a TPE scene unit evaluation module, an environmental scene simulation and deduction module, and a simulation prediction process configuration module.

[0101] The TPE scene unit construction module is connected to the TPE scene unit evaluation module, the TPE scene unit evaluation module is connected to the environmental scene simulation and deduction module, and the environmental scene simulation and deduction module is connected to the simulation prediction process configuration module.

[0102] The TPE scenario unit building module is used by the industrial digital management platform to read targeted data from raw heterogeneous data, attach it to the platform's pre-built basic management objects according to the targeted time slice, and update the basic management objects in the pre-built TPE scenario units.

[0103] The TPE scene unit evaluation module is used by the industrial digital management platform to calculate the geopolitical environment volatility for geopolitical environment scenarios under targeted time slices, customize adaptive aggregation mode to aggregate targeted data, and evaluate the channel exposure vector and node structure feature parameters of each TPE scene unit after aggregation.

[0104] The environmental scenario simulation and deduction module is used to conduct integrated simulation and deduction of geopolitical environmental scenarios on the industrial digital management platform based on basic statistical characteristics, geopolitical environmental volatility, channel exposure vectors, and node structural characteristic parameters.

[0105] The simulation prediction process configuration module is used to filter TPE scenario records in the TPE scenario unit. The filtered TPE scenario records are input into the DEA of the industrial digital management platform. The DEA decision unit outputs the corresponding industrial input matrix and industrial output matrix. Combined with the simulation simulation process, the simulation prediction process of the geopolitical environment scenario is configured to complete the industrial digital management.

[0106] The above content is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A method for industrial digital management integrating geopolitical environmental data, characterized in that... This includes: the industrial digital management platform reads targeted data from raw heterogeneous data and attaches it to the platform's pre-built basic management objects according to the targeted time slices, and updates the basic management objects in the pre-built TPE scenario units; the industrial digital management platform calculates geopolitical environment volatility for geopolitical environment scenarios under the targeted time slices, customizes an adaptive aggregation mode to aggregate targeted data, and evaluates the channel exposure vector and node structure feature parameters of each TPE scenario unit after aggregation; based on basic statistical features, geopolitical environment volatility, channel exposure vector, and node structure feature parameters, the industrial digital management platform jointly conducts integrated geopolitical environment scenario simulation and deduction; TPE scenario records are filtered in the TPE scenario units, and the filtered TPE scenario records are input into the DEA of the industrial digital management platform. The DEA decision unit outputs the corresponding industrial input matrix and industrial output matrix, and configures the simulation prediction process of geopolitical environment scenarios in conjunction with the simulation and deduction process to complete industrial digital management.

2. The industrial digital management method integrating geopolitical environmental data according to claim 1, characterized in that... The specific pre-construction process of the basic management objects and the TPE scenario units is as follows: A basic management object table is established in the data layer of the industrial digital management platform. The basic management objects specifically include: a regional management object table (R table), a management tool object table (T table), a technology path object table (P table), a geo-environment scenario object table (E table), an action channel object table (C table), and an on-chain node object table (N table). Based on the regional management object table (R table), management tool object table (T table), technology path object table (P table), and geo-environment scenario object table (E table), the industrial digital management platform, under a set time slice identifier (time_slice), traverses each region according to the regional identifier (R_ID), filters out the management tool identifier (T_ID) and geo-environment scenario identifier (E_ID) that are effective for the targeted region and its technology path stage within that time slice, and combines this with the technology path stage identifier (P_ID) and its associated on-chain node set to generate a corresponding TPE scenario unit record for each set of (R_ID, T_ID, P_ID, E_ID, time_slice) that meets the conditions. Target indicator fields, channel exposure vector fields, and node structure feature fields are reserved in the scenario unit.

3. The industrial digital management method integrating geopolitical environmental data according to claim 2, characterized in that... The industrial digital management platform reads time-stamped targeted setting data from raw heterogeneous data, performs time standardization processing on the targeted setting data, generates hierarchical time dimensions for the targeted setting data, and links the targeted setting data to basic management objects according to the hierarchical time dimensions, so that the targeted setting data are associated with the corresponding regional management objects, management tool objects, technology path objects, and geo-environment scene objects respectively. The industrial digital management platform updates the pre-constructed TPE scene unit records according to the combination of regional identifiers, management tool identifiers, technology path stage identifiers, geo-environment scene identifiers, and time slice identifiers, so that each TPE scene unit record carries targeted setting data that matches the management object and time dimension under the targeted time slice.

4. The industrial digital management method integrating geopolitical environmental data according to claim 1, characterized in that... The process involves calculating geopolitical volatility for a geopolitical scenario under a targeted time slice and using a customized adaptive aggregation mode to aggregate targeted data. Specifically, the analysis process is as follows: Geopolitical data sequences for each targeted time slice are read from the original heterogeneous data. The industrial digital management platform calculates the corresponding geopolitical volatility based on the geopolitical data sequences within the current and adjacent targeted time slices. The geopolitical volatility of the scenario is compared with a predefined geopolitical volatility threshold. If the geopolitical volatility is less than or equal to the threshold, the aggregation mode is set to the first aggregation mode; otherwise, it is set to the second aggregation mode. Based on the time aggregation window corresponding to the aggregation mode, the targeted data within that window is time-aggregated. Basic statistical features are then calculated for the time-aggregated targeted data.

5. The industrial digital management method integrating geopolitical environmental data according to claim 4, characterized in that... The industrial digital management platform, for each TPE scenario unit, carries a regional identifier, management tool identifier, technology path stage identifier, geographical environment scenario identifier, and time slice identifier. It filters all data records whose timestamps fall within the time aggregation window from the targeted data, and groups the filtered data records according to the targeted settings. Based on the grouped data records, it calculates basic statistical characteristics for various targeted indicators of each TPE scenario unit within the time aggregation window. These basic statistical characteristics include one or more of the following: mean, standard deviation, maximum value, minimum value, linear trend representation, and number of anomalies for each type of targeted indicator of the TPE scenario unit. The basic statistical characteristics are then written into the targeted indicator field of the corresponding TPE scenario unit.

6. The industrial digital management method integrating geopolitical environmental data according to claim 1, characterized in that... The specific analysis process for evaluating the channel exposure vector and node structure feature parameters of each TPE scenario unit is as follows: The industrial digital management platform calculates the structural weight of each function channel under the TPE scenario unit based on the preset mapping relationship between the geo-environment scenario object and the function channel object table, combined with the association relationship between the on-chain node object table and the technology path object table. Environmental response feature parameters are constructed based on basic statistical characteristics, and dynamic feature coefficients are obtained through matching these parameters. The industrial digital management platform uses the product of the structural weight of each function channel and the corresponding dynamic adjustment coefficient as the channel exposure vector of that function channel under the TPE scenario unit, and writes it into the channel exposure vector field of the TPE scenario unit. Based on the on-chain node set covered by the TPE scenario unit, the industrial digital management platform extracts the structural information of each node from the on-chain node object table, statistically analyzes structural feature indicators, combines these indicators to form node structure feature parameters, and writes them into the node structure feature field of the TPE scenario unit.

7. The industrial digital management method integrating geopolitical environmental data according to claim 1, characterized in that... The simulation management layer of the industrial digital management platform receives the target TPE scene unit selected by the user. Based on the target TPE scene unit, the simulation management layer loads the corresponding geopolitical environment scene simulation process template. It uses the basic statistical features, geopolitical environment volatility, channel exposure vector, and node structure feature parameters already written in the target TPE scene unit as simulation input features to dynamically correct the geopolitical environment scene simulation process template. The corrected geopolitical environment scene simulation process template simulates and extrapolates the change trajectory of the target indicators of the target TPE scene unit through preset geopolitical environment scenario combinations and time evolution constraints.

8. The industrial digital management method integrating geopolitical environmental data according to claim 1, characterized in that... The DEA decision unit outputs the corresponding industry input matrix and industry output matrix, and combines the simulation and deduction process to configure the simulation prediction process of the geopolitical environment scenario. The specific analysis process is as follows: The simulation management layer of the industrial digital management platform receives the filtering conditions set by the user, including preset regional conditions, time segment conditions, and geopolitical environment scenario constraints. It filters out several TPE scenario records from each TPE scenario record and records them as a sample set. The sample set is used as the input of the DEA decision unit, and the industry input matrix and industry output matrix of the sample set are output. The industry input matrix and industry output matrix are combined and configured with the simulation input features. According to the geopolitical environment scenario combination and time evolution constraints, the upper limit of industrial resource input, the output target range, the geopolitical environment disturbance intensity, and the transmission coefficient of each action channel are set in an integrated manner to complete the simulation prediction configuration process of the geopolitical environment scenario.

9. The industrial digital management method integrating geopolitical environmental data according to claim 8, characterized in that... The simulation management layer of the industrial digital management platform uses each input element column in the industrial input matrix as the initial parameters for industrial resource allocation boundary and input intensity in the simulation prediction process; and uses each output index column in the industrial output matrix as the target output parameter and performance evaluation threshold in the simulation prediction process.

10. An industrial digital management system integrating geopolitical environmental data, employing the industrial digital management method integrating geopolitical environmental data as described in any one of claims 1-9, characterized in that... The system includes: a TPE scenario unit construction module, used by the industrial digital management platform to read targeted data from raw heterogeneous data and attach it to the platform's pre-built basic management objects according to the targeted time slice, and update the basic management objects in the pre-built TPE scenario units; a TPE scenario unit evaluation module, used by the industrial digital management platform to calculate geopolitical environment volatility for geopolitical environment scenarios under the targeted time slice, customize an adaptive aggregation mode to aggregate targeted data, and evaluate the channel exposure vector and node structure feature parameters of each TPE scenario unit after aggregation; an environmental scenario simulation and deduction module, used to conduct integrated geopolitical environment scenario simulation and deduction on the industrial digital management platform based on basic statistical features, geopolitical environment volatility, channel exposure vector, and node structure feature parameters; and a simulation prediction process configuration module, used to filter TPE scenario records in TPE scenario units, input the filtered TPE scenario records into the DEA of the industrial digital management platform, the DEA decision unit outputs the corresponding industrial input matrix and industrial output matrix, and configures the simulation prediction process of geopolitical environment scenarios in conjunction with the simulation and deduction process to complete industrial digital management.

Images