A digital economic park operation situation analysis system

Abstract

This invention relates to the field of digital operation and management of industrial parks, and discloses a digital economy park operation status analysis system; comprising: a multi-source data access module, an operation evidence event generation module, an input time lag calibration module, a status analysis window backfilling module, a multi-layer operation graph construction module, a dynamic baseline generation module, a conflict evidence adjudication module, and a status tracing and output module. The multi-source data access module accesses enterprise business activity data, digital resource usage data, spatial carrier usage data, industrial chain collaboration data, and service item data; the operation evidence event generation module converts the accessed data into operation evidence events; the input time lag calibration module configures time lag labels; the status analysis window backfilling module backfills delayed or supplementary events to the corresponding window; and the multi-layer operation graph construction module establishes the relationships between enterprise entities, spatial carriers, industrial chain nodes, digital resources, and service items.

Description

Technical Field

[0001] This invention relates to the field of digital operation and management of industrial parks, specifically to a digital economy industrial park operation status analysis system. Background Technology

[0002] With the clustering and development of the digital economy industry, an increasing number of industrial parks are accommodating companies engaged in software development, industrial internet, data services, platform operation, cloud services, and digital content. The operational activities of these companies are reflected not only in traditional data such as occupancy area, number of employees, energy consumption, and service transaction records, but also in various other data points including contract delivery, online platform activities, cloud resource usage, network bandwidth consumption, supply chain collaboration, and policy service response. Therefore, the park operation management system needs to comprehensively analyze multi-source data to understand the overall operational status of the park, changes in supply chain nodes, the activity level of enterprise clusters, the occupancy status of spatial carriers, and the status of service transactions.

[0003] Existing park operation analysis systems typically rely on basic enterprise information, building energy consumption, investment attraction data, service work orders, security data, or operational statistics. They generate park status results through indicator aggregation, fixed threshold judgment, model prediction, or large-screen display. While this approach can handle routine data statistics and display, for digital economy parks, the data generation and storage times of different business systems often differ. For example, contract and order data may be synchronized daily or weekly, service item data may be updated according to processing milestones, cloud resource usage data may be collected hourly, and supply chain collaboration data may require subsequent supplementary entry. Directly calculating statistics based on data storage time can easily include historical business activities in the current statistical window, leading to a discrepancy between the park's operational status and the actual business cycle.

[0004] Meanwhile, digital economy enterprises are characterized by a high online fulfillment rate, relatively stable office space occupancy, high flexibility in digital resource utilization, and complex supply chain collaboration relationships. Existing systems, when conducting situational analysis, typically use changes in energy consumption, orders, service delays, or resource usage as single indicators, making it difficult to distinguish between changes in enterprise operational activity, abnormal spatial carriers, supply chain disruptions, and changes in service response. For example, when an enterprise's cloud resource usage increases while office energy consumption decreases, traditional systems may identify this as an operational anomaly; similarly, when energy consumption in the same building is abnormal but the operational data of resident enterprises does not change synchronously, it may be misjudged as fluctuations in the operations of a group of enterprises. Therefore, existing technologies still require a digital economy park operational situational analysis system capable of performing time-lag calibration on multi-source asynchronous data, graphically correlating operational evidence, and adjudicating and causally tracing conflicting evidence. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a digital economy park operation status analysis system to solve the technical problems existing in the prior art.

[0006] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0007] A digital economy park operation status analysis system, comprising:

[0008] The system includes a multi-source data access module, an operational evidence event generation module, an inbound time delay calibration module, a situation analysis window backfilling module, a multi-layer operational map construction module, a dynamic baseline generation module, a conflict evidence adjudication module, and a situation tracing output module.

[0009] The multi-source data access module is used to access data on business activities of enterprises in the digital economy park, data on the use of digital resources, data on the use of spatial carriers, data on industrial chain collaboration, and data on park service items, and to configure the data source type, event occurrence time, and event entry time for the accessed data.

[0010] The operational evidence event generation module is used to convert the data accessed by the multi-source data access module into operational evidence events. Operational evidence events include event subject identifier, event type, event occurrence time, event entry time, data source type, event intensity value, associated object identifier, and initial credibility.

[0011] The data entry time lag calibration module is used to determine the regular delay threshold and supplementary entry judgment threshold for the corresponding data source type based on the data entry delay records of the same data source type in the historical statistical period, and to configure real-time tags, delayed tags or supplementary entry tags for operational evidence events based on the time difference between the event occurrence time and the event data entry time.

[0012] The situation analysis window backfilling module is used to generate a continuously arranged situation analysis window according to a preset statistical granularity. It determines the situation analysis window to which the operational evidence event belongs based on the event occurrence time of the operational evidence event, and writes the operational evidence events configured with lag tags or supplementary tags into the situation analysis window to which their event occurrence time belongs.

[0013] The multi-layer operation graph construction module is used to construct a multi-layer operation graph including the enterprise entity layer, spatial carrier layer, industrial chain node layer, digital resource layer, and service item layer, and to attach operation evidence events to the corresponding nodes or edges in the multi-layer operation graph according to the event entity identifier and the associated object identifier.

[0014] The dynamic baseline generation module is used to group enterprises according to their industry chain node, park entry stage, and scale range. It generates a dynamic baseline based on the intensity value of operational evidence events in the historical situation analysis window within the same group, and determines the direction and intensity of deviation of operational evidence events in the current situation analysis window relative to the dynamic baseline.

[0015] The conflict evidence adjudication module is used to determine the corresponding situational attribution type and adjudication credibility when operational evidence events corresponding to the same event subject identifier or the same associated object identifier have different deviation directions, based on the sequential dependency between event types, the deviation direction of adjacent nodes in the multi-layer operation graph, the time delay label of the operational evidence event, the initial credibility, and historical consistency.

[0016] The situation attribution output module is used to output the operational situation results of the digital economy park based on the deviation direction, deviation intensity, situation attribution type, and adjudication credibility, and output the operational evidence event chain corresponding to the operational situation results in chronological order of event occurrence.

[0017] Preferably, the multi-source data access module includes an operational activity data access unit, a digital resource data access unit, a spatial carrier data access unit, an industrial chain collaboration data access unit, and a service item data access unit;

[0018] The business activity data access unit accesses enterprise contract order data, project delivery data, R&D project data, and job recruitment data;

[0019] The digital resource data access unit accesses enterprise cloud resource usage data, network bandwidth usage data, and platform account activity data.

[0020] The spatial carrier data access unit accesses data on the buildings, floors, rooms, workstations, and energy consumption of the carriers where the enterprise is located.

[0021] The supply chain collaborative data access unit accesses supply and demand collaboration data, joint project data, and upstream and downstream related data among enterprises;

[0022] The service item data access unit accesses data on park policy applications, financing matchmaking, talent services, site changes, and project filing.

[0023] Preferably, the operational evidence event generation module includes a field standardization unit, a subject mapping unit, and an event type configuration unit;

[0024] The field standardization unit converts the accessed data into a unified field format, which includes data record identifier, data source type, original subject name, subject certificate identifier, event occurrence time, event entry time, event value field, and associated object field.

[0025] The entity mapping unit generates the event entity identifier corresponding to the enterprise entity based on the original entity name, entity certificate identifier, platform account identifier, spatial carrier identifier, and service item application entity identifier.

[0026] The event type configuration unit configures the accessed data as order growth events, order decline events, resource usage enhancement events, resource usage decline events, carrier occupancy change events, industry collaboration change events, service item addition events, or service item delay events according to the data source type and event value field.

[0027] Preferably, the warehousing time delay calibration module includes a delay record extraction unit, a time delay threshold configuration unit, and a tag configuration unit;

[0028] The delayed record extraction unit extracts the delayed entry records between the event entry time and the event occurrence time of each data record of the same data source type within the historical statistical period;

[0029] The time delay threshold configuration unit configures the median value of the inbound delay records as the regular delay threshold for the corresponding data source type, and configures the upper quartile value of the inbound delay records as the supplementary entry judgment threshold for the corresponding data source type, according to the distribution of the inbound delay records.

[0030] When the time difference between the event entry time and the event occurrence time of the operational evidence event does not exceed the conventional delay threshold of the corresponding data source type, the tag configuration unit configures a real-time tag for the operational evidence event.

[0031] When the time difference exceeds the normal delay threshold but does not exceed the supplementary recording judgment threshold, a lag label is configured for the operational evidence event; when the time difference exceeds the supplementary recording judgment threshold, a supplementary recording label is configured for the operational evidence event.

[0032] Preferably, the situation analysis window backfilling module includes a window division unit, an event attribution unit, and a history window update unit;

[0033] The window partitioning unit generates a situation analysis window with a window number, window start time, and window end time according to a preset statistical granularity.

[0034] The event attribution unit assigns operational evidence events whose occurrence time falls between the window start time and the window end time to the corresponding situation analysis window.

[0035] When an operational evidence event is configured with a lag tag or a supplementary tag, the historical window update unit writes the operational evidence event into the situation analysis window to which the event occurred, and records the update batch, update data source type, update event quantity, and backfill event identifier of the situation analysis window.

[0036] Preferably, the multi-layer operation graph construction module includes a node generation unit and an edge generation unit;

[0037] The node generation unit generates enterprise entity nodes, spatial carrier nodes, industrial chain nodes, digital resource nodes, and service item nodes;

[0038] The edge generation unit generates the following: entry edge between enterprise main node and spatial carrier node; affiliation edge between enterprise main node and industrial chain node; usage edge between enterprise main node and digital resource node; processing edge between enterprise main node and service item node; collaboration edge between enterprise main nodes; and upstream and downstream edge between industrial chain nodes.

[0039] The multi-layer operation graph construction module attaches operation evidence events to enterprise entity nodes based on event subject identifiers, and attaches operation evidence events to spatial carrier nodes, industrial chain nodes, digital resource nodes, service item nodes, or corresponding edges based on associated object identifiers.

[0040] Preferably, the dynamic baseline generation module includes a grouping unit, a baseline calculation unit, and a deviation determination unit;

[0041] The grouping unit groups the enterprises according to their industry chain node, park entry stage, and scale range;

[0042] The baseline calculation unit extracts the operational evidence event intensity values ​​of the enterprise entities within the same group in multiple consecutive historical situation analysis windows, and generates the dynamic baseline corresponding to the group based on the mean and standard deviation of the operational evidence event intensity values. The dynamic baseline includes the baseline center value, upper boundary, and lower boundary. The baseline center value is the mean, the upper boundary is determined based on the sum of the mean and the standard deviation, and the lower boundary is determined based on the difference between the mean and the standard deviation.

[0043] The deviation determination unit compares the operational evidence event intensity value of the enterprise entity within the current situation analysis window with the dynamic baseline of its group. When the operational evidence event intensity value is higher than the upper boundary, the deviation direction is determined as a positive deviation, and the deviation intensity is determined based on the difference between the operational evidence event intensity value and the upper boundary.

[0044] When the operational evidence event strength value is lower than the lower boundary, the deviation direction is determined as a negative deviation, and the deviation strength is determined based on the difference between the lower boundary and the operational evidence event strength value.

[0045] When the operational evidence event strength value is between the upper and lower boundaries, the deviation direction is determined as no deviation.

[0046] Preferably, the conflict evidence adjudication module includes a conflict identification unit, an adjacency verification unit, a time delay priority processing unit, a historical consistency calculation unit, and an adjudication unit;

[0047] The conflict identification unit identifies operational evidence events where the same event subject identifier or the same associated object identifier deviates in opposite directions within the same situation analysis window.

[0048] The adjacency verification unit extracts the adjacent nodes in the multi-layer operation graph that are connected to the nodes corresponding to the event subject identifier or the associated object identifier, and reads the operational evidence event deviation direction of the adjacent nodes within the same situation analysis window;

[0049] When the time-delay priority processing unit has operational evidence events with delayed or supplementary labels in operational evidence events that deviate in the opposite direction, the situation analysis window backfilling module first performs backfilling on the operational evidence event, and then the adjudication unit makes an adjudication.

[0050] The historical consistency calculation unit generates historical consistency based on the percentage of times the deviation direction and corresponding situation attribution type of the same event type in multiple consecutive historical situation analysis windows for the same event subject identifier or the same associated object identifier.

[0051] The adjudication unit determines the situational attribution type and adjudication credibility corresponding to the conflict based on the event type, time delay label, initial credibility, deviation direction of adjacent nodes, and historical consistency of the operational evidence event with the opposite deviation direction.

[0052] Preferably, the adjudication unit determines the situation attribution type according to the following rules:

[0053] When both the order growth event or project delivery event and the resource utilization enhancement event for the same business entity are positively deviating, and the energy consumption event for the corresponding business entity is negatively deviating, the situation attribution type is determined as a change in business activity.

[0054] If the number of business entities whose business activity events have the same deviation direction among the enterprise entities associated with the same spatial carrier node does not reach the same-direction ratio threshold, and the carrier energy consumption event corresponding to the spatial carrier node forms a same-direction deviation, the situation attribution type will be determined as spatial carrier anomaly.

[0055] When an upstream enterprise experiences a decline in orders, and a downstream enterprise experiences a decline in project delivery or service delays, the cause of the situation will be classified as a change in supply chain collaboration.

[0056] When both the service item delay event and the order decline event or resource usage decline event corresponding to the same enterprise entity are negatively deviated, the situation attribution type is determined to be service response change.

[0057] Preferably, the situation tracing output module includes a situation result generation unit, an evidence chain generation unit, and a supplementary evidence generation unit;

[0058] The situational results generation unit generates operational situational results according to the overall park, industrial chain nodes, spatial carriers, enterprise groups, and service item categories.

[0059] The evidence chain generation unit generates an operational evidence event chain according to the chronological order of events, including operational evidence events, data source types, time delay labels, situation analysis window numbers, backfill event identifiers, deviation directions, deviation intensities, situation attribution types, and adjudication credibility.

[0060] When the conflict evidence adjudication module has not determined the situational attribution type, the supplementary evidence generation unit reads the set of event types involved in the conflict adjudication, determines the missing event type according to the preset event type combination table, and writes the data source type corresponding to the missing event type into the supplementary evidence data source list.

[0061] In summary, the present invention has the following main beneficial effects:

[0062] By uniformly converting enterprise operational data, digital resource usage data, spatial carrier usage data, industrial chain collaboration data, and park service data into operational evidence events, and recording the event occurrence time, event entry time, data source type, event intensity value, and associated object identifier in each operational evidence event, and configuring real-time tags, lagging tags, and supplementary tags based on historical entry delay records of the same data source type, and backfilling the operational evidence events corresponding to the lagging tags or supplementary tags into the situation analysis window to which the event occurrence time belongs, the system avoids the error of historical operational facts being incorrectly included in the current statistical window due to different update cycles of different business systems, different data supplementation times, or different entry delays. This allows the park's operational situation analysis to be collected according to the actual occurrence time of business facts, reducing situation mismatch and misjudgment caused by multi-source asynchronous data in digital economy parks.

[0063] By constructing a multi-layered operational map including enterprise entity layer, spatial carrier layer, industrial chain node layer, digital resource layer, and service item layer, and attaching operational evidence events to corresponding nodes or edges according to event entity identifiers and related object identifiers, and then generating dynamic baselines according to the industrial chain node to which the enterprise entity belongs, the stage of entry into the park, and the scale range, deviation judgments are made on the event intensity values ​​under the same group and the same basic event type. This achieves the effect of hierarchical correlation analysis of changes in enterprise operation, changes in digital resource usage, changes in spatial carrier, changes in industrial chain collaboration, and changes in service item handling. It avoids relying solely on fixed indicator thresholds or total data of a single park for judgment, enabling digital economy enterprises with light assets, online operations, and flexible resource usage characteristics to conduct situational comparisons under the same industry stage and scale conditions.

[0064] When operational evidence events corresponding to the same event subject identifier or the same associated object identifier deviate in different directions, the system combines the sequential dependencies between event types, the deviation direction of adjacent nodes in the multi-layered operational graph, the time lag label of operational evidence events, initial credibility, and historical consistency to adjudicate conflicting evidence. It then outputs the situational attribution type, adjudication credibility, and an operational evidence event chain sorted by event occurrence time, achieving the effect of traceable attribution of the source of situational changes. When the evidence is insufficient to determine the situational attribution type, the system also generates a list of data sources to be supplemented based on the event type combination table. This avoids mixing energy consumption anomalies, service delays, supply chain disturbances, or enterprise operational fluctuations, improving the interpretability, verifiability, and targeted nature of subsequent supplementary evidence processing of the park's operational situation results. Attached Figure Description

[0065] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0066] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Example 1

[0068] refer to Figure 1 A digital economy park operation status analysis system, comprising:

[0069] The system includes a multi-source data access module, an operational evidence event generation module, an inbound time delay calibration module, a situation analysis window backfilling module, a multi-layer operational map construction module, a dynamic baseline generation module, a conflict evidence adjudication module, and a situation tracing output module.

[0070] The multi-source data access module is used to access data on business activities of enterprises in the digital economy park, data on the use of digital resources, data on the use of spatial carriers, data on industrial chain collaboration, and data on park service items, and to configure the data source type, event occurrence time, and event entry time for the accessed data.

[0071] The operational evidence event generation module is used to convert the data accessed by the multi-source data access module into operational evidence events. Operational evidence events include event subject identifier, event type, event occurrence time, event entry time, data source type, event intensity value, associated object identifier, and initial credibility.

[0072] The data entry time lag calibration module is used to determine the regular delay threshold and supplementary entry judgment threshold for the corresponding data source type based on the data entry delay records of the same data source type in the historical statistical period, and to configure real-time tags, delayed tags or supplementary entry tags for operational evidence events based on the time difference between the event occurrence time and the event data entry time.

[0073] The situation analysis window backfilling module is used to generate a continuously arranged situation analysis window according to a preset statistical granularity. It determines the situation analysis window to which the operational evidence event belongs based on the event occurrence time of the operational evidence event, and writes the operational evidence events configured with lag tags or supplementary tags into the situation analysis window to which their event occurrence time belongs.

[0074] The multi-layer operation graph construction module is used to construct a multi-layer operation graph including the enterprise entity layer, spatial carrier layer, industrial chain node layer, digital resource layer, and service item layer, and to attach operation evidence events to the corresponding nodes or edges in the multi-layer operation graph according to the event entity identifier and the associated object identifier.

[0075] The dynamic baseline generation module is used to group enterprises according to their industry chain node, park entry stage, and scale range. It generates a dynamic baseline based on the intensity value of operational evidence events in the historical situation analysis window within the same group, and determines the direction and intensity of deviation of operational evidence events in the current situation analysis window relative to the dynamic baseline.

[0076] The conflict evidence adjudication module is used to determine the corresponding situational attribution type and adjudication credibility when operational evidence events corresponding to the same event subject identifier or the same associated object identifier have different deviation directions, based on the sequential dependency between event types, the deviation direction of adjacent nodes in the multi-layer operation graph, the time delay label of the operational evidence event, the initial credibility, and historical consistency.

[0077] The situation attribution output module is used to output the operational situation results of the digital economy park based on the deviation direction, deviation intensity, situation attribution type, and adjudication credibility, and output the operational evidence event chain corresponding to the operational situation results in chronological order of event occurrence.

[0078] This system can be deployed on a park operation management server, a park data platform server, or a cloud computing server. The system includes a processor, memory, a database, and data interface services. The memory stores program instructions that can be executed by the processor. When the processor executes these program instructions, it implements the functions of a multi-source data access module, an operation evidence event generation module, an input time delay calibration module, a situation analysis window backfilling module, a multi-layer operation map construction module, a dynamic baseline generation module, a conflict evidence adjudication module, and a situation tracing output module.

[0079] The database includes tables for enterprise entities, data source types, operational evidence events, inventory delay records, situation analysis windows, multi-level operational graphs, dynamic baselines, conflict resolution records, event type combinations, system parameters, and operational evidence events. The Enterprise Entity Table records the unified entity identifier of enterprises in the park; the Data Source Type Table records the data sources of different business systems; the Operational Evidence Event Table records operational evidence events formed by the conversion of multi-source data; the Storage Delay Record Table records the delay between the event occurrence time and the event storage time; the Situation Analysis Window Table records the window number, window start time, and window end time; the Multi-Level Operational Map Table records the nodes and edges between enterprise entities, spatial carriers, industry chain nodes, digital resources, and service items; the Dynamic Baseline Table records the baseline center value, upper boundary, and lower boundary corresponding to each enterprise group and each event type; the Conflict Adjudication Record Table records the set of conflict events, adjudication basis, situation attribution type, and adjudication credibility; the Event Type Combination Table records the event type combinations required for different situation attribution types; the System Parameter Table records the number of historical windows, initial threshold, same-direction ratio threshold, credibility weight, and output threshold; and the Operational Evidence Event Chain List records the evidence chain in the process of forming situation results.

[0080] In this embodiment, a digital economy park refers to an industrial park where software development, industrial internet, digital content, data services, platform operation, cloud services, digital marketing, or digital service delivery activities take place. The system does not rely on a single operational indicator as the basis for situational assessment. Instead, it converts data on enterprise operations, digital resource usage, spatial carrier usage, supply chain collaboration, and park service items generated during park operations into operational evidence events. This data is then combined with the event occurrence time, event entry time, event intensity value, graph correlation, and historical window data for situational analysis.

[0081] The multi-source data access module is used to access existing and authorized data within the digital economy park. Access methods include API calls, database synchronization, message queue reception, or batch file import. Regardless of the access method used, each data record is configured with the data source type, event occurrence time, and event entry time.

[0082] Enterprise operational data includes enterprise contract and order data, project delivery data, R&D project data, and job posting data. This type of data can be sourced from the park's enterprise service platform, contract filing system, project delivery filing system, R&D project management system, and recruitment service system, and must include at least the enterprise name, main business license, contract or project number, project delivery milestone, R&D project number, job category, event occurrence time, and event entry time into the database.

[0083] Digital resource usage data includes enterprise cloud resource usage data, network bandwidth usage data, and platform account activity data. This type of data can originate from park cloud resource management platforms, network egress management systems, public computing power platforms, software service platforms, or platform account management systems, and at least includes platform account identifiers, enterprise entity identifiers or mappable fields, cloud resource instance identifiers, bandwidth usage records, storage usage records, platform login or call records, resource usage occurrence time, and event entry time.

[0084] Space usage data includes data on the buildings, floors, rooms, workstations, and energy consumption of businesses. This data can be sourced from the park's property management system, building management system, workstation management system, and energy consumption management system, and includes at least the building number, floor number, room number, workstation number, business occupancy record, energy consumption record, space change time, and event entry time.

[0085] Supply chain collaboration data includes data on supply and demand cooperation between enterprises, joint project data, and upstream and downstream relationship data. This type of data can come from the park's supply chain management system, enterprise supply and demand matching system, joint project registration system, and upstream and downstream relationship maintenance system, and at least includes enterprise entity identifier, collaborating enterprise identifier, supply chain node identifier, supply and demand relationship type, joint project identifier, time of occurrence of collaboration relationship, and time of event entry into the database.

[0086] The data on park services includes processing data for park policy applications, financing matchmaking, talent services, site changes, and project filings. This type of data can be sourced from the park's government service system, policy application system, financing matchmaking system, talent service system, site change system, and project filing system, and should at least include the service item number, applicant, item category, item status, acceptance time, completion time, node update time, and event entry time.

[0087] The event occurrence time refers to the actual time when the corresponding business fact is formed, such as the contract signing time, project delivery milestone completion time, resource usage record collection time, building occupancy change time, service item milestone update time, or service item completion time. The event entry time refers to the time when the corresponding data record enters the system's database or the park's data platform. The system records both the event occurrence time and the event entry time separately, and does not use the event entry time to replace the event occurrence time.

[0088] When a data record lacks an event occurrence time, the system does not directly use that data record for backfilling the situation analysis window. If the event occurrence time can be determined based on the business node time, then the corresponding business node time is used as the event occurrence time, and a time estimation identifier is recorded in the operational evidence event table; if the event occurrence time cannot be determined, then the data record is written to the abnormal data table, and the supplementary certificate generation unit writes the corresponding data source type into the supplementary certificate data source list.

[0089] The operational evidence event generation module includes a field standardization unit, a subject mapping unit, and an event type configuration unit.

[0090] The field standardization unit converts the incoming data into a unified field format. This unified field format includes data record identifier, data source type, original entity name, entity certificate identifier, event occurrence time, event entry time, event value field, and associated object field. The event value field records order amount, order quantity, delivery quantity, resource usage, number of workstations, energy consumption value, and processing time or node dwell time. The associated object field records building, industry chain node, digital resource account, service item number, or collaborating enterprise identifier.

[0091] The entity mapping unit generates an event entity identifier for the corresponding enterprise entity based on the original entity name, entity certificate identifier, platform account identifier, space carrier identifier, and service item application entity identifier. When different names of the same enterprise exist in different data sources, the entity mapping unit prioritizes using the entity certificate identifier for mapping; when the entity certificate identifier is missing, mapping is performed according to the binding relationship between the enterprise name, platform account identifier, space carrier entry record, and service item application entity. The entity mapping result is written to the enterprise entity table, which includes at least the event entity identifier, entity certificate identifier, standard entity name, platform account identifier, space carrier identifier, and service item application entity identifier.

[0092] The event type configuration unit adopts a two-stage event type configuration approach. In the first stage, basic operational evidence events are generated based on the data source type and event value fields. These basic event types include order-related events, resource usage events, carrier occupancy events, industry collaboration events, and service item events. In the second stage, after the dynamic baseline generation module determines the deviation direction, the basic event types are combined with the deviation direction to generate deviation event types for conflict evidence adjudication. Specifically, order-related events are combined with positive deviations to form order growth events, and with negative deviations to form order decline events; resource usage events are combined with positive deviations to form resource usage enhancement events, and with negative deviations to form resource usage decline events; service item events are combined with negative deviations to form service item delay events; carrier occupancy events are combined with their corresponding deviation directions to form carrier occupancy change events; and industry collaboration events are combined with their corresponding deviation directions to form industry collaboration change events.

[0093] By configuring event types in two stages, the operational evidence event generation module can first generate basic operational evidence events that do not depend on the dynamic baseline. After the dynamic baseline generation module completes the deviation judgment, deviation event types such as order growth event, order decline event, resource utilization enhancement event, and resource utilization decline event are generated, so that the execution order of the modules remains consistent.

[0094] Each operational evidence event includes the event subject identifier, event type, event occurrence time, event entry time, data source type, event intensity value, associated object identifier, and initial credibility. The event intensity value is obtained from the event value field based on the basic event type. The event intensity value for order-related events can be the number of contract orders, contract amount, or number of project delivery nodes; the event intensity value for resource usage-related events can be the number of cloud resource instances, computing resource usage, bandwidth usage, storage usage, or number of platform account activities; the event intensity value for space carrier-related events can be the number of workstations, room area, building energy consumption value, or number of carrier changes; the event intensity value for service item-related events can be the number of newly added items, the number of unresolved items, processing time, or node dwell time.

[0095] The dynamic baseline is calculated only within the same enterprise group and the same basic event type, and does not mix event intensity values ​​of different dimensions such as order amount, resource usage, energy consumption value, and event processing time.

[0096] Initial credibility is used to record the credible foundation for operational evidence events before they enter the conflict evidence adjudication module. Initial credibility is determined based on field completeness, subject mapping method, event occurrence time validity, and data source interface stability. Field completeness indicates whether the data record contains fields for data record identifier, data source type, event subject information, event occurrence time, event entry time, and event value. Subject mapping method indicates whether the event subject identifier is directly determined by the subject's certificate identifier, or indirectly determined by the company name, platform account, spatial carrier identifier, and service application subject. Event occurrence time validity indicates whether the event occurrence time is an original field or inferred from a business node time. Data source interface stability indicates whether the corresponding data source continuously generates valid records within the historical statistical period. All value rules are stored in the system parameter table. Pre-set rules can be used during initial deployment, and can be updated based on manually verified records after the system is running.

[0097] The data entry time lag calibration module includes a delay record extraction unit, a time lag threshold configuration unit, and a tag configuration unit. This module addresses the issues of different data update cycles, supplementary entry times, and data entry delays across different park business systems. Instead of directly calculating operational status statistics based on event entry time, the system performs time lag calibration based on the difference between the event occurrence time and the event entry time.

[0098] For the For each operational evidence event, the delay in its entry into the warehouse is expressed as follows:

[0099] ;

[0100] in, Indicates the first Delay in the entry of operational evidence into the database; Indicates the first The time when the operational evidence event was entered into the database; Indicates the first The time of occurrence of the operational evidence incident.

[0101] The delayed record extraction unit extracts delayed entry records for each data record of the same data source type within the historical statistical period. The historical statistical period is configured by the system parameter table and can be set to the most recent situation analysis windows, the most recent calendar month, the most recent quarter, or a custom operational statistical period for the park. The system calculates delayed entry records separately according to the same data source type and does not mix delayed entry records from different data source types in the calculation.

[0102] For data source type The standard delay threshold and the supplementary recording judgment threshold are expressed as follows:

[0103] ;

[0104] ;

[0105] in, Indicates the data source type The standard delay threshold; Indicates the data source type The threshold for supplementary recording; median( ) represents the median value function; Represents the upper quartile value function; Indicates the first The data source type for each operational evidence event. The median value is used as the normal delay threshold to characterize the normal data entry speed for this data source type; the upper quartile value is used as the supplementary entry judgment threshold to distinguish between normal delayed entry and supplementary entry exceeding the normal fluctuation range.

[0106] When the number of delayed inbound records of the same data source type within a historical statistical period is less than the preset sample size configured in the system parameter table, the time lag threshold configuration unit uses the initial regular delay threshold and initial supplementary record judgment threshold for that data source type in the system parameter table. The initial regular delay threshold and initial supplementary record judgment threshold are configured by the park management terminal according to the data update cycle of the corresponding business system. When the number of subsequent delayed inbound records reaches the preset sample size, the time lag threshold configuration unit updates the threshold based on the aforementioned median and upper quartile values.

[0107] The tag configuration unit configures time-delay tags based on the relationship between the entry delay of operational evidence events and the regular delay threshold and supplementary entry judgment threshold for the corresponding data source type. At that time, configure real-time tags for the operational evidence event; when At that time, a delayed label is configured for the operational evidence event; when At that time, a supplementary entry tag is configured for the operational evidence event. Real-time tags, delayed tags, and supplementary entry tags are all written to the operational evidence event table.

[0108] The situation analysis window backfilling module includes window division units, event attribution units, and history window update units.

[0109] The window division units generate a continuous arrangement of situation analysis windows according to a preset statistical granularity. The preset statistical granularity is configured by the system parameter table and can be set to daily, weekly, ten-day, monthly, or a custom operating cycle for the park. Each situation analysis window includes a window number, a window start time, and a window end time.

[0110] The event attribution unit assigns operational evidence events whose occurrence time falls between the window start and end times to the corresponding situation analysis window based on the event occurrence time. The event attribution unit does not use the event entry time as the basis for window attribution.

[0111] When an operational evidence event is configured with a delayed or supplementary tag, the historical window update unit writes the operational evidence event into the situation analysis window to which its occurrence time belongs, and records the update batch, update data source type, update event quantity, and backfill event identifier for that situation analysis window. The update batch records how many times the situation analysis window has been backfilled; the update data source type records the data source that triggered the backfill; the update event quantity records the number of operational evidence events involved in this backfill; and the backfill event identifier indicates in the evidence chain that the event is not counted in the current window according to the event entry time, but rather backfilled to the historical window to which the event occurrence time belongs.

[0112] After the historical window update unit completes backfilling, the system re-triggers the dynamic baseline comparison, deviation direction judgment, deviation intensity calculation, conflict evidence adjudication, and operational evidence event chain update of the affected situation analysis window. If the backfilling event changes the historical situation result, the situation tracing output module records the backfilling event identifier, the window number before backfilling, the window number after backfilling, and the update batch in the operational evidence event chain.

[0113] The multi-layered operation graph construction module includes node generation units and edge generation units. The multi-layered operation graph includes the enterprise entity layer, spatial carrier layer, industrial chain node layer, digital resource layer, and service item layer.

[0114] The node generation unit generates enterprise entity nodes, spatial carrier nodes, industry chain nodes, digital resource nodes, and service item nodes. The enterprise entity node identifier is generated from the event entity identifier; the spatial carrier node identifier is generated from the building number, floor number, room number, or workstation number; the industry chain node identifier is generated from the park's industry chain classification table; the digital resource node identifier is generated from the cloud resource instance identifier, platform account identifier, or network resource identifier; and the service item node identifier is generated from the service item number or item category.

[0115] The edge generation unit generates entry edges between enterprise entity nodes and spatial carrier nodes, ownership edges between enterprise entity nodes and industry chain nodes, usage edges between enterprise entity nodes and digital resource nodes, processing edges between enterprise entity nodes and service item nodes, collaboration edges between enterprise entity nodes, and upstream and downstream edges between industry chain nodes. Entry edges record the association between enterprises and buildings, floors, rooms, or workstations; ownership edges record the industry chain nodes to which enterprises belong; usage edges record the usage relationships between enterprises and cloud resources, network resources, or platform accounts; processing edges record the application, acceptance, transfer, completion, or delay relationships between enterprises and service items; collaboration edges record the supply and demand, joint projects, or collaborative delivery relationships between enterprises; and upstream and downstream edges record the upstream and downstream connection relationships between industry chain nodes.

[0116] The multi-layered operation graph construction module attaches operation evidence events to enterprise entity nodes based on event subject identifiers, and attaches them to spatial carrier nodes, industry chain nodes, digital resource nodes, service item nodes, or corresponding edges based on associated object identifiers. For example, order-related events are attached to the enterprise entity node and the industry chain node to which the enterprise belongs; resource usage-related events are attached to the enterprise entity node and digital resource nodes; service item-related events are attached to the enterprise entity node and service item nodes; carrier occupancy-related events are attached to spatial carrier nodes; and industry collaboration-related events are attached to collaboration edges between enterprise entities or upstream and downstream edges between industry chain nodes.

[0117] The dynamic baseline generation module includes a grouping unit, a baseline calculation unit, and a deviation determination unit.

[0118] The grouping unit categorizes enterprises based on their industry chain node, park entry stage, and scale range. Industry chain nodes are derived from the industry chain node layer in the multi-layered operation map; the park entry stage is determined by the enterprise's entry time, the time of handling its first service request, or the time of its first space occupancy; the scale range is determined by the enterprise's scale field, number of workstations, number of projects, or the scale field of authorized access in the park's records. The grouping results are written into the enterprise grouping table.

[0119] The baseline calculation unit extracts the operational evidence event intensity values ​​of enterprise entities within the same group across multiple consecutive historical situation analysis windows, and generates a dynamic baseline corresponding to that group based on the mean and standard deviation of the operational evidence event intensity values. (Number of historical situation analysis windows) Configured by the system parameter table. When the number of history windows is insufficient... When the number of historical windows reaches a certain threshold, the system calculates a temporary dynamic baseline using existing historical windows and marks this temporary dynamic baseline as a low-sample baseline; when the number of historical windows reaches a certain threshold... After that, the system is updated to a formal dynamic baseline.

[0120] For grouping and basic event types Its dynamic baseline is represented as:

[0121] ;

[0122] ;

[0123] ;

[0124] in, Indicates grouping Basic event types The baseline center value; Indicates grouping Basic event types Standard deviation; Indicates grouping In the Basic event types within the historical situation analysis window The event intensity value; This indicates the number of historical situation analysis windows involved in the calculation; Indicates the upper boundary of the dynamic baseline; This indicates the lower boundary of the dynamic baseline.

[0125] The deviation determination unit compares the operational evidence event strength value of the enterprise entity within the current situation analysis window with the dynamic baseline of its group. For the first... An operational evidence event, its event strength value is The group to which it belongs is The basic event type is .when When the deviation direction is determined to be positive, then the deviation is considered positive. When the deviation direction is determined to be negative, then... At this point, the deviation direction is determined to be no deviation. The deviation intensity is expressed as:

[0126] ;

[0127] in, Indicates the first The degree of deviation of the operational evidence event; Indicates the first The event intensity value of each operational evidence event; Indicates grouping and basic event types The upper boundary of the corresponding dynamic baseline; Indicates grouping and basic event types The lower boundary of the corresponding dynamic baseline.

[0128] Dynamic baselines are calculated only within the same basic event type and the same enterprise group. For example, order-related events are compared only with historical intensity values ​​of order-related events, resource usage-related events are compared only with historical intensity values ​​of resource usage-related events, and service-related events are compared only with historical intensity values ​​of service-related events; cross-event type mixed calculations are not performed.

[0129] The conflict evidence adjudication module includes a conflict identification unit, an adjacency verification unit, a time delay priority processing unit, a historical consistency calculation unit, and an adjudication unit.

[0130] The conflict identification unit identifies operational evidence events with the same event subject identifier or the same associated object identifier that deviate in opposite directions within the same situation analysis window. Opposite deviation directions mean that within the operational evidence events corresponding to the same event subject identifier or the same associated object identifier, both positive and negative deviation events exist simultaneously. The set of conflict events is written into the conflict resolution record table.

[0131] The adjacency verification unit extracts adjacent nodes connected to the nodes corresponding to the event subject identifier or associated object identifier in the multi-layer operation graph, and reads the deviation direction of the operational evidence event within the same situational analysis window. Adjacency nodes include spatial carrier nodes, industry chain nodes, digital resource nodes, and service item nodes connected to the same enterprise entity, as well as upstream and downstream industry chain nodes connected to the same industry chain node and other enterprise entity nodes connected to the same spatial carrier node. The adjacency verification unit counts the number of positive deviations, negative deviations, and no deviations among adjacent nodes and generates adjacency deviation statistics results.

[0132] When an operational evidence event with a delayed or supplementary tag exists in an operational evidence event that deviates in the opposite direction, the time lag priority processing unit first performs backfilling on the operational evidence event by the situation analysis window backfilling module, and then the adjudication unit makes an adjudication. This process is used to prevent late-entry or supplementary data from directly participating in the current window adjudication, which could cause the situation attribution to deviate from the actual occurrence cycle of the event.

[0133] The historical consistency calculation unit generates historical consistency based on the percentage of corresponding occurrences of deviation directions and corresponding situational attribution types for the same event type combination within multiple consecutive historical situational analysis windows, using the same event subject identifier or the same associated object identifier. The same event type combination refers to a set of basic event types participating in the same conflict resolution within the same situational analysis window, and whose deviation directions are identical. For example, a positive deviation of order-related events, a positive deviation of resource usage-related events, and a negative deviation of carrier occupancy-related events constitute a single event type combination.

[0134] Historical consistency is represented as: in, Indicates the type of situational attribution Corresponding historical consistency; This indicates that within multiple consecutive historical situation analysis windows, combinations of the same event types are classified as situation attribution types. The number of times; This indicates the total number of times the same event type combination occurs within multiple consecutive historical situation analysis windows. At this time, the system uses the initial historical consistency value configured in the system parameter table; the initial historical consistency value can be set to a neutral value or formed from the park's historical manual verification records. When the system generates an actual adjudication record, the historical consistency calculation unit updates it using the actual historical window record. .

[0135] The adjudication unit determines the situational attribution type and adjudication credibility corresponding to the conflict based on the event type, time delay label, initial credibility, deviation direction of adjacent nodes, and historical consistency of the operational evidence events with opposite deviation directions. Adjudication credibility indicates the degree of evidentiary support for the adjudication result. For candidate situational attribution types... The credibility of its ruling is expressed as follows:

[0136] ;

[0137] in, Indicates candidate situation attribution type The credibility of the ruling; This represents the average initial credibility of the operational evidence events that participated in the adjudication of this candidate situation attribution type. The time-delay label credibility factor represents the operational evidence event that participated in the adjudication of the candidate situation attribution type. Indicates candidate situation attribution type Corresponding historical consistency; Indicates the attribution type of candidate situations for adjacent nodes in the graph. Consistency support value; , , , These are the weighting coefficients recorded in the system parameter table, and .

[0138] in, The average of the initial credibility of the operational evidence events involved in the adjudication is obtained. The real-time label, the delayed label, and the supplementary label are converted from time-delay labels. They correspond to different credibility factors, and the values ​​of the credibility factors are configured by the system parameter table. Generated by the historical consistency calculation unit; This is generated by the consistency ratio of neighboring nodes, which is the proportion of neighboring nodes whose deviation direction is consistent with the candidate situation attribution type out of the total number of neighboring nodes. During initial deployment... , , , Equal weighting can be used; after the system is running, adjustments can be accurately recorded based on manual review results or historical rulings.

[0139] The adjudication unit determines the situation attribution type according to the following rules.

[0140] When both the order growth event or project delivery event and the resource utilization enhancement event for the same business entity deviate positively, and the energy consumption event for the corresponding business entity deviates negatively, the adjudication unit will determine the situation attribution type as "change in business activity." This rule applies to situations where digital economy enterprises experience increased online fulfillment and cloud resource usage, but office space energy consumption does not increase simultaneously.

[0141] When the number of business entities associated with the same spatial carrier node whose business activity events show a common deviation direction does not reach the same-direction proportion threshold, and the carrier energy consumption events corresponding to that spatial carrier node show a common-direction deviation, the adjudication unit will determine the situation attribution type as spatial carrier anomaly. The same-direction proportion threshold is a threshold used to determine whether the business activities of business entities under the same spatial carrier node have formed a group of same-direction changes. This threshold is stored in the system parameter table and can be preset by the park management terminal or determined based on the historical proportion of same-direction changes in business activity events of the same spatial carrier node in multiple consecutive historical situation analysis windows.

[0142] When an upstream supply chain node experiences a decline in orders, and a downstream supply chain node experiences a decline in project delivery or service delays, the adjudication unit will determine the situation attribution type as a supply chain collaborative change. This rule requires that there are upstream and downstream edges in a multi-layered operational graph between upstream and downstream supply chain nodes, and that the corresponding events belong to the same situation analysis window, or, after backfilling, belong to the same situation analysis window.

[0143] When both the service item delay event and the order decline event or resource usage decline event corresponding to the same enterprise entity deviate negatively, the adjudication unit will determine the situation attribution type as service response change. The basis for the occurrence of a service item delay event is the result of a negative deviation between the duration of the service item node in the acceptance, review, feedback, or completion status and the corresponding dynamic baseline, specifically based on the direction and intensity of the deviation for that type of service item event.

[0144] If multiple situational attribution types simultaneously meet the adjudication rules, the adjudication unit calculates the adjudication credibility of each candidate situational attribution type and selects the situational attribution type with the highest adjudication credibility as the adjudication result of the current conflict; if the adjudication credibility of all candidate situational attribution types is lower than the output threshold in the system parameter table, the final situational attribution type is not output, and the supplementary evidence generation unit generates a list of data sources to be supplemented.

[0145] The situational tracing output module includes a situational result generation unit, an evidence chain generation unit, and a supplementary evidence generation unit.

[0146] The situational assessment results generation unit generates operational situational assessment results separately for the overall park, industry chain nodes, spatial carriers, enterprise groups, and service item categories. The overall park operational situational assessment results are formed by summarizing the situational assessment results for each industry chain node, spatial carrier, enterprise group, and service item category; the operational situational assessment results for each industry chain node are formed based on operational evidence events of the enterprise entities under that industry chain node, upstream and downstream node events, and industry collaboration change events; the operational situational assessment results for each spatial carrier are formed based on enterprise entity events, carrier energy consumption events, and occupancy change events under the spatial carrier nodes; the operational situational assessment results for each enterprise group are formed based on the direction and intensity of deviation of the enterprise entities within the same group from the dynamic baseline; and the operational situational assessment results for each service item category are formed based on new service item events, service item delay events, and completed service item events.

[0147] The evidence chain generation unit generates an operational evidence event chain based on the chronological order of events, including operational evidence events, data source type, time lag label, situation analysis window number, backfill event identifier, deviation direction, deviation intensity, situation attribution type, and decision credibility. The operational evidence event chain is not sorted by event entry time. If an event has undergone backfilling, the evidence chain records the backfill event identifier and the situation analysis window number after backfilling.

[0148] When the conflict evidence adjudication module has not determined the situational attribution type, the supplementary evidence generation unit reads the set of event types involved in the conflict adjudication, determines the missing event type according to the preset event type combination table, and writes the data source type corresponding to the missing event type into the supplementary evidence data source list.

[0149] The event type combination table includes situational attribution types, necessary event types, optional event types, corresponding data source types, and missing event handling rules. Necessary event types corresponding to changes in operational activity include operational activity events and digital resource events, with corresponding data source types including enterprise operational activity data and digital resource usage data. Necessary event types corresponding to spatial carrier anomalies include spatial carrier events and related enterprise operational activity events, with corresponding data source types including spatial carrier usage data and enterprise operational activity data. Necessary event types corresponding to changes in supply chain collaboration include upstream operational activity events and downstream delivery events, with corresponding data source types including supply chain collaboration data and enterprise operational activity data. Necessary event types corresponding to changes in service response include service item delay events, as well as enterprise operational activity events or resource usage events, with corresponding data source types including park service item data, enterprise operational activity data, or digital resource usage data.

[0150] The supplementary certificate generation unit determines the missing necessary event types in the current conflict resolution based on the event type combination table. For example, when a business entity only has abnormal carrier energy consumption events but lacks business activity events and digital resource events, the supplementary certificate generation unit writes the business activity data and digital resource usage data into the supplementary certificate data source list; when a node in the industrial chain has a downstream delivery decline event but lacks upstream business activity events, the supplementary certificate generation unit writes the industrial chain collaboration data or upstream enterprise business activity data into the supplementary certificate data source list.

[0151] The system first uses a multi-source data access module to access data on enterprise operations, digital resource usage, spatial carrier usage, industry chain collaboration, and park service items. A field standardization unit converts data from different sources into a unified field format. A subject mapping unit generates event subject identifiers based on subject license identifiers, platform account identifiers, spatial carrier identifiers, and service item application subject identifiers. Finally, an event type configuration unit generates basic operational evidence events.

[0152] Subsequently, the data entry delay calibration module calculates the data entry delay for each operational evidence event, generates a regular delay threshold and a supplementary entry judgment threshold based on historical data entry delay records of the same data source type, and configures real-time, delayed, or supplementary entry tags for operational evidence events. The situation analysis window backfilling module assigns operational evidence events to the corresponding situation analysis windows according to their occurrence time; for delayed or supplementary entry tag events, the system writes the situation analysis window to which the event occurrence time belongs and records the backfilled event identifier.

[0153] The multi-layered operation map construction module generates a multi-layered operation map based on the enterprise entity, spatial carrier, industry chain nodes, digital resources, and service items, and attaches operation evidence events to corresponding nodes or edges. The dynamic baseline generation module groups enterprises according to industry chain nodes, entry stage, and scale range. It generates a dynamic baseline based on the event intensity values ​​of the same group and the same basic event type within a continuous historical trend analysis window, and determines the deviation direction and deviation intensity of each operation evidence event in the current window. Based on the basic event type and deviation direction, the system further generates deviation event types such as order growth events, order decline events, resource utilization enhancement events, resource utilization decline events, and service item delay events.

[0154] When the system detects operational evidence events with opposite deviation directions within the same enterprise entity, spatial carrier node, or industrial chain node, the conflict evidence adjudication module is activated. The conflict identification unit determines the set of conflict events; the adjacency verification unit reads the deviation directions of relevant adjacent nodes in the multi-layer operational graph; the time lag priority processing unit performs window backfilling for events with lagging labels and supplementary labels; the historical consistency calculation unit calculates the proportion of attribution-corresponding times for combinations of the same event type in the historical window; and the adjudication unit, combining event type, time lag label, initial credibility, adjacent node deviation direction, and historical consistency, determines the situational attribution type and adjudication credibility.

[0155] Finally, the situation attribution output module outputs operational situation results for the entire park, industry chain nodes, spatial carriers, enterprise groups, and service item categories, and generates an operational evidence event chain sorted by event occurrence time. If the adjudication unit fails to determine the situation attribution type, the supplementary evidence generation unit generates a list of data sources to be supplemented based on the event type combination table.

[0156] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A digital economy park operation status analysis system, characterized in that, include: The system includes a multi-source data access module, an operational evidence event generation module, an inbound time delay calibration module, a situation analysis window backfilling module, a multi-layer operational map construction module, a dynamic baseline generation module, a conflict evidence adjudication module, and a situation tracing output module. The multi-source data access module is used to access data on business activities of enterprises in the digital economy park, data on the use of digital resources, data on the use of spatial carriers, data on industrial chain collaboration, and data on park service items, and to configure the data source type, event occurrence time, and event entry time for the accessed data. The operational evidence event generation module is used to convert the data accessed by the multi-source data access module into operational evidence events. Operational evidence events include event subject identifier, event type, event occurrence time, event entry time, data source type, event intensity value, associated object identifier, and initial credibility. The data entry time lag calibration module is used to determine the regular delay threshold and supplementary entry judgment threshold for the corresponding data source type based on the data entry delay records of the same data source type in the historical statistical period, and to configure real-time tags, delayed tags or supplementary entry tags for operational evidence events based on the time difference between the event occurrence time and the event data entry time. The situation analysis window backfilling module is used to generate a continuously arranged situation analysis window according to a preset statistical granularity. It determines the situation analysis window to which the operational evidence event belongs based on the event occurrence time of the operational evidence event, and writes the operational evidence events configured with lag tags or supplementary tags into the situation analysis window to which their event occurrence time belongs. The multi-layer operation graph construction module is used to construct a multi-layer operation graph including the enterprise entity layer, spatial carrier layer, industrial chain node layer, digital resource layer, and service item layer, and to attach operation evidence events to the corresponding nodes or edges in the multi-layer operation graph according to the event entity identifier and the associated object identifier. The dynamic baseline generation module is used to group enterprises according to their industry chain node, park entry stage, and scale range. It generates a dynamic baseline based on the intensity value of operational evidence events in the historical situation analysis window within the same group, and determines the direction and intensity of deviation of operational evidence events in the current situation analysis window relative to the dynamic baseline. The conflict evidence adjudication module is used to determine the corresponding situational attribution type and adjudication credibility when operational evidence events corresponding to the same event subject identifier or the same associated object identifier have different deviation directions, based on the sequential dependency between event types, the deviation direction of adjacent nodes in the multi-layer operation graph, the time delay label of the operational evidence event, the initial credibility, and historical consistency. The situation attribution output module is used to output the operational situation results of the digital economy park based on the deviation direction, deviation intensity, situation attribution type, and adjudication credibility, and output the operational evidence event chain corresponding to the operational situation results in chronological order of event occurrence.

2. The digital economy park operation status analysis system according to claim 1, characterized in that, The multi-source data access module includes a business activity data access unit, a digital resource data access unit, a spatial carrier data access unit, an industry chain collaboration data access unit, and a service item data access unit; The business activity data access unit accesses enterprise contract order data, project delivery data, R&D project data, and job recruitment data; The digital resource data access unit accesses enterprise cloud resource usage data, network bandwidth usage data, and platform account activity data. The spatial carrier data access unit accesses data on the buildings, floors, rooms, workstations, and energy consumption of the carriers where the enterprise is located. The supply chain collaborative data access unit accesses supply and demand collaboration data, joint project data, and upstream and downstream related data among enterprises; The service item data access unit accesses data on park policy applications, financing matchmaking, talent services, site changes, and project filing.

3. The digital economy park operation status analysis system according to claim 2, characterized in that, The operational evidence event generation module includes a field standardization unit, a subject mapping unit, and an event type configuration unit; The field standardization unit converts the accessed data into a unified field format, which includes data record identifier, data source type, original subject name, subject certificate identifier, event occurrence time, event entry time, event value field, and associated object field. The entity mapping unit generates the event entity identifier corresponding to the enterprise entity based on the original entity name, entity certificate identifier, platform account identifier, spatial carrier identifier, and service item application entity identifier. The event type configuration unit configures the accessed data as order growth events, order decline events, resource usage enhancement events, resource usage decline events, carrier occupancy change events, industry collaboration change events, service item addition events, or service item delay events according to the data source type and event value field.

4. The digital economy park operation status analysis system according to claim 3, characterized in that, The inbound time delay calibration module includes a delay record extraction unit, a time delay threshold configuration unit, and a tag configuration unit; The delayed record extraction unit extracts the delayed entry records between the event entry time and the event occurrence time of each data record of the same data source type within the historical statistical period; The time delay threshold configuration unit configures the median value of the inbound delay records as the regular delay threshold for the corresponding data source type, and configures the upper quartile value of the inbound delay records as the supplementary entry judgment threshold for the corresponding data source type, according to the distribution of the inbound delay records. When the time difference between the event entry time and the event occurrence time of the operational evidence event does not exceed the conventional delay threshold of the corresponding data source type, the tag configuration unit configures a real-time tag for the operational evidence event. When the time difference exceeds the normal delay threshold but does not exceed the supplementary recording judgment threshold, a lag label is configured for the operational evidence event; When the time difference exceeds the threshold for supplementary recording, a supplementary recording label is configured for the operational evidence event.

5. The digital economy park operation status analysis system according to claim 4, characterized in that, The situation analysis window backfilling module includes a window division unit, an event attribution unit, and a history window update unit; The window partitioning unit generates a situation analysis window with a window number, window start time, and window end time according to a preset statistical granularity. The event attribution unit assigns operational evidence events whose occurrence time falls between the window start time and the window end time to the corresponding situation analysis window. When an operational evidence event is configured with a lag tag or a supplementary tag, the historical window update unit writes the operational evidence event into the situation analysis window to which the event occurred, and records the update batch, update data source type, update event quantity, and backfill event identifier of the situation analysis window.

6. The digital economy park operation status analysis system according to claim 5, characterized in that, The multi-layer operation graph construction module includes a node generation unit and an edge generation unit; The node generation unit generates enterprise entity nodes, spatial carrier nodes, industrial chain nodes, digital resource nodes, and service item nodes; The edge generation unit generates the following: entry edge between enterprise main node and spatial carrier node; affiliation edge between enterprise main node and industrial chain node; usage edge between enterprise main node and digital resource node; processing edge between enterprise main node and service item node; collaboration edge between enterprise main nodes; and upstream and downstream edge between industrial chain nodes. The multi-layer operation graph construction module attaches operation evidence events to enterprise entity nodes based on event subject identifiers, and attaches operation evidence events to spatial carrier nodes, industrial chain nodes, digital resource nodes, service item nodes, or corresponding edges based on associated object identifiers.

7. The digital economy park operation status analysis system according to claim 6, characterized in that, The dynamic baseline generation module includes a grouping unit, a baseline calculation unit, and a deviation determination unit; The grouping unit groups the enterprises according to their industry chain node, park entry stage, and scale range; The baseline calculation unit extracts the operational evidence event intensity values ​​of the enterprise entities within the same group in multiple consecutive historical situation analysis windows, and generates the dynamic baseline corresponding to the group based on the mean and standard deviation of the operational evidence event intensity values. The dynamic baseline includes the baseline center value, upper boundary, and lower boundary. The baseline center value is the mean, the upper boundary is determined based on the sum of the mean and the standard deviation, and the lower boundary is determined based on the difference between the mean and the standard deviation. The deviation determination unit compares the operational evidence event intensity value of the enterprise entity within the current situation analysis window with the dynamic baseline of its group. When the operational evidence event intensity value is higher than the upper boundary, the deviation direction is determined as a positive deviation, and the deviation intensity is determined based on the difference between the operational evidence event intensity value and the upper boundary. When the operational evidence event strength value is lower than the lower boundary, the deviation direction is determined as a negative deviation, and the deviation strength is determined based on the difference between the lower boundary and the operational evidence event strength value. When the operational evidence event strength value is between the upper and lower boundaries, the deviation direction is determined as no deviation.

8. The digital economy park operation status analysis system according to claim 7, characterized in that, The conflict evidence adjudication module includes a conflict identification unit, an adjacency verification unit, a time delay priority processing unit, a historical consistency calculation unit, and an adjudication unit. The conflict identification unit identifies operational evidence events where the same event subject identifier or the same associated object identifier deviates in opposite directions within the same situation analysis window. The adjacency verification unit extracts the adjacent nodes in the multi-layer operation graph that are connected to the nodes corresponding to the event subject identifier or the associated object identifier, and reads the operational evidence event deviation direction of the adjacent nodes within the same situation analysis window; When the time-delay priority processing unit has operational evidence events with delayed or supplementary tags in operational evidence events that deviate in the opposite direction, the situation analysis window backfilling module first performs backfilling on the operational evidence event, and then the adjudication unit makes an adjudication. The historical consistency calculation unit generates historical consistency based on the percentage of times the deviation direction and corresponding situation attribution type of the same event type in multiple consecutive historical situation analysis windows for the same event subject identifier or the same associated object identifier. The adjudication unit determines the situational attribution type and adjudication credibility corresponding to the conflict based on the event type, time delay label, initial credibility, deviation direction of adjacent nodes, and historical consistency of the operational evidence event with the opposite deviation direction.

9. The digital economy park operation status analysis system according to claim 8, characterized in that, The adjudication unit determines the situation attribution type according to the following rules: When both the order growth event or project delivery event and the resource utilization enhancement event for the same business entity are positively deviating, and the energy consumption event for the carrier corresponding to that business entity is negatively deviating, the situation attribution type will be determined as a change in business activity. If the number of business entities whose business activity events have the same deviation direction does not reach the same-direction ratio threshold, and the energy consumption events of the corresponding space carrier node form a same-direction deviation, the situation attribution type will be determined as space carrier anomaly. When an upstream enterprise experiences a decline in orders, and a downstream enterprise experiences a decline in project delivery or service delays, the cause of the situation will be classified as a change in supply chain collaboration. When both the service item delay event and the order decline event or resource usage decline event corresponding to the same enterprise entity are negatively deviated, the situation attribution type is determined to be service response change.

10. The digital economy park operation status analysis system according to claim 9, characterized in that, The situation tracing output module includes a situation result generation unit, an evidence chain generation unit, and a supplementary evidence generation unit. The situational results generation unit generates operational situational results according to the overall park, industrial chain nodes, spatial carriers, enterprise groups, and service item categories. The evidence chain generation unit generates an operational evidence event chain according to the chronological order of events, including operational evidence events, data source types, time delay labels, situation analysis window numbers, backfill event identifiers, deviation directions, deviation intensities, situation attribution types, and adjudication credibility. When the conflict evidence adjudication module has not determined the situational attribution type, the supplementary evidence generation unit reads the set of event types involved in the conflict adjudication, determines the missing event type according to the preset event type combination table, and writes the data source type corresponding to the missing event type into the supplementary evidence data source list.

Images