Digital supervision system and method for highway water transportation construction market

Abstract

The invention discloses a digital supervision system and method for a road water transportation construction market, and relates to the technical field of market digital supervision, and the system comprises a market authentication archiving module which is used for building globally unique digital identities for all participants, and constructing a dynamically linked credible archive library; the internet-of-things perception inspection module is used for extracting project information and personnel identity information in the credible archive library, identifying violation events of the highway water transportation engineering in combination with internet-of-things monitoring data acquired by internet-of-things equipment, and outputting structured engineering construction records; and the engineering twinborn tracking module is used for constructing a digital twinborn body fused with highway and water transportation engineering projects and outputting progress quality linkage analysis information in combination with the engineering projects and engineering construction records. According to the invention, multiple modules are integrated for cooperative work, so that the intelligent level of engineering supervision is improved, a scientific decision basis is provided for managers, resource allocation is optimized on the whole, and the probability of occurrence of engineering delay and quality accidents is reduced.

Description

Technical Field

[0001] This invention relates to the field of digital market supervision technology, and in particular to a digital supervision system and method for the highway and waterway construction market. Background Technology

[0002] The highway and waterway construction market encompasses the construction of large-scale transportation infrastructure such as roads, bridges, waterways, and ports. Projects are typically characterized by large investment scales, long construction periods, numerous participating entities, and wide geographical distribution. During project implementation, traditional quality and safety supervision and management mainly rely on manual on-site inspections, paper-based document circulation, and post-event reporting. This not only involves complex business processes and large amounts of data but also makes it difficult to achieve effective, real-time, and dynamic supervision throughout the entire process, easily leading to problems such as delayed information transmission, low collaboration efficiency, and blind spots in supervision.

[0003] Against this backdrop, digital supervision has emerged as a new governance model. Its core lies in utilizing next-generation information technologies such as big data, the Internet of Things (IoT), artificial intelligence (AI), and BeiDou navigation to achieve full-chain digital management of project participants, construction processes, quality and safety, and credit performance. Currently, some digital supervision systems are able to integrate industry data, build unified data resource databases, and enable visualized querying and verification of project information through graphical interfaces. Some systems have also connected to IoT monitoring devices, enabling digital supervision of project sites and personnel safety, and timely issuance of early warning information. Digital supervision is gradually evolving from single-stage management to comprehensive collaboration covering the entire "construction, management, maintenance, and operation" chain and all participating parties, aiming to achieve simultaneous data archiving upon generation and build a full lifecycle digital archive.

[0004] However, existing technologies also face some common problems in their development. On the one hand, some existing systems have inconsistent standards, leading to barriers to data sharing and business collaboration, making it difficult to achieve networked operation and an integrated service experience. On the other hand, some digital upgrade projects exhibit a phenomenon of "emphasizing construction over maintenance, hardware over software, and technology over mechanisms," failing to fully realize their economies of scale and continuous operational efficiency. Furthermore, when applying digital systems, challenges may arise from collaboration barriers caused by the failure to update existing organizational and business rules in sync, as well as challenges related to data security and privacy protection.

[0005] Therefore, developing a digital regulatory system that can cover the entire chain of the highway and waterway construction market and realize closed-loop management from project access and process supervision to credit evaluation is an important technical path to address the above-mentioned industry pain points and improve industry governance capabilities. Summary of the Invention

[0006] Therefore, it is necessary to provide a digital monitoring system and method for the highway and waterway construction market to address the aforementioned technical issues.

[0007] In a first aspect, the present invention provides a digital monitoring system for the highway and waterway construction market, the system comprising:

[0008] The market authentication and archiving module is used to establish globally unique digital identities for all participants and to build a dynamically linked trusted archive.

[0009] The IoT sensing and inspection module is used to extract project information and personnel identity information from the trusted archive, and combine it with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records.

[0010] The engineering twin tracking module is used to build a digital twin of highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records.

[0011] The risk simulation decision-making module is used to simulate future engineering processes by importing environmental forecast data into the digital twin based on schedule and quality linkage analysis information, and output a simulation analysis report of the current engineering project.

[0012] The credit assessment and payment module is used to build smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

[0013] Furthermore, the digital oversight system also includes:

[0014] The visualization decision-making module is used to build a configurable visualization interface through graphical interaction, and supports the linkage of graphs and data and data drill-down functions to realize human interaction.

[0015] Open extension integration modules are used to build connection channels for external ecosystems. Through standardized interfaces and security protocols, they provide application plugins and analysis tools for development, testing, and the platform.

[0016] The security management and protection module is used to deploy network security protection tools, monitor and issue early warnings for network indicators in real time, and formulate and implement data backup and disaster recovery strategies.

[0017] Furthermore, the IoT sensing and inspection module includes:

[0018] The multi-source data fusion module is used to establish a data mapping model, identify and associate the relationships between projects, personnel and equipment from different data sources, and output standardized qualification data to extract basic project information, personnel identity data and performance relationships from the trusted archive.

[0019] The heterogeneous device acquisition module is used to identify the types of IoT devices connected, add three-dimensional coordinates and timestamps to the IoT monitoring data collected by IoT devices, and establish spatial association with qualification data;

[0020] The violation identification and analysis module is used to integrate qualification data and IoT monitoring data, establish a violation identification and analysis model based on the preset violation patterns of highway and waterway engineering, verify the legality of personnel and equipment, and output an event record set containing confidence scores by combining confidence verification and filtering.

[0021] The record generation and distribution module is used to integrate the identification results of the event record set with other non-abnormal event sets, encapsulate them into structured engineering construction records, and automatically select push channels according to the violation type and urgency of the event record set, and push them to the designated terminal through an encrypted channel.

[0022] Furthermore, by integrating qualification data and IoT monitoring data, a violation identification and analysis model is established based on pre-defined violation patterns in highway and waterway engineering. This model verifies the legality of personnel and equipment, and, combined with confidence level verification and filtering, outputs a set of event records containing confidence scores, including:

[0023] By using semantic alignment technology for engineering projects, qualification data and IoT monitoring data are synchronized to obtain a real-time data stream that is uniformly aligned in terms of time, space and semantics. Dynamic permission verification is used to analyze the behavior patterns of personnel and equipment in the current engineering project and output the permission verification results.

[0024] Based on the project graph-driven violation identification and analysis model, the system integrates permission verification results and IoT monitoring data. By reasoning the pre-defined highway and waterway violation patterns through the graph, it identifies violation events in the current project and outputs an asynchronous event list.

[0025] Based on an asynchronous event list, confidence levels are calibrated for violations in various engineering projects. Probability scores are calculated by combining environmental factors, and an event record set that meets the confidence threshold requirements is output.

[0026] Furthermore, based on the project graph-driven violation identification and analysis model, which integrates permission verification results and IoT monitoring data, and uses graph inference to pre-defined highway and waterway violation patterns, it identifies abnormal behaviors in the current project and outputs a list of asynchronous events, including:

[0027] Match entities in the current real-time data stream with nodes in the knowledge graph of highway and waterway engineering projects to identify the engineering project where personnel and equipment are currently located.

[0028] Based on the current engineering project, the pre-set monitoring targets in the knowledge graph are activated, and the cross-modal attention mechanism is used to extract the device behavior features and environmental operation parameters from the database, and align them with the permission verification results in a unified vector space to obtain a multimodal feature vector.

[0029] Traverse the violation pattern paths starting from the current project node, use graph neural networks to calculate the matching degree between real-time data and the pre-defined violation patterns in the graph, and obtain the suspected event set;

[0030] Based on the suspected event set, the temporal logic of the entire construction operation cycle is analyzed by tracing back the scene state changes of the preceding and subsequent frames to make temporal logic judgments on the suspected events, and output the violation events that have been verified by the temporal logic; and the violation events are encapsulated into structured data objects and merged to generate a structured asynchronous event list.

[0031] Furthermore, based on the asynchronous event list, confidence levels are calibrated for asynchronous events in various engineering projects. Probability scores are calculated by combining environmental factors, and the output set of event records that meet the confidence threshold requirements includes:

[0032] The asynchronous event list is parsed and preprocessed, and the events are classified according to different project types based on the project identifier to generate an event dataset grouped by project.

[0033] Based on the timestamp and spatial coordinates of each event, the corresponding environmental factor data is dynamically extracted and normalized to form an environmental factor feature vector.

[0034] A prior probability distribution is established based on historical data, and a posterior probability is calculated by combining real-time environmental factors. Different weight coefficients are configured for different engineering projects, and the post-calibration confidence of each event is calculated using Bayes' formula to generate a time probability matrix containing the post-calibration confidence score.

[0035] The confidence threshold is dynamically adjusted based on the event type and project. Combined with a multi-level filtering strategy, violations below the preset threshold are filtered out. The retained violations are subject to risk assessment and classification, and an event record set is generated through structured encapsulation.

[0036] Furthermore, the engineering twin tracking module includes:

[0037] The twin construction module is used to build a three-dimensional digital twin based on real-time data streams and building information models, realize dynamic visualization and entity mapping of engineering projects, and automatically adjust the corresponding texture attributes in the digital twin when violations are detected in the engineering construction records.

[0038] The status tracking and detection module is used to track the real-time status of personnel and equipment in the digital twin, and integrate violations in the engineering construction records to analyze the potential impact of violations on the construction progress.

[0039] The schedule and quality linkage module is used to integrate schedule data, quality inspection indicators, and engineering construction records to perform multi-dimensional linkage analysis and output schedule and quality linkage analysis information.

[0040] Furthermore, the real-time status of personnel and equipment is tracked within the digital twin, and violations in the construction records are integrated to analyze their potential impact on construction progress, including:

[0041] Multi-target tracking algorithms are used to continuously track the trajectories of people and equipment in a digital twin, and behavioral pattern recognition is performed by combining video stream data to obtain the group operation mode.

[0042] By deeply linking violations in the construction records with real-time entity behavior, an event behavior association rule base is established. The correlation between events and behaviors is calculated through graph neural networks to identify the scope of the impact of violations on the construction process.

[0043] Based on the results of event behavior correlation analysis and combined with schedule data, the potential impact of violations on construction progress is assessed. By quantifying the delay time of individual violations, multi-scenario impact simulations are conducted to output risk quantification indicators for violations.

[0044] Furthermore, by integrating progress data, quality inspection indicators, and project construction records, a multi-dimensional linkage analysis is conducted, outputting progress-quality linkage analysis information including:

[0045] The system acquires the current project's progress data and quality inspection indicators, aligns them with the project construction records in time and space to form a standardized data stream, and dynamically adjusts the weight coefficients of quality and progress according to the project stage to construct a schedule-quality coupling analysis model and output a schedule-quality correlation matrix.

[0046] By mapping violations in the project construction records to the schedule and quality correlation matrix, and using graph neural networks to analyze the causal relationships of the event chain, the causes of schedule delays or quality problems are analyzed, and schedule and quality linkage analysis information is obtained.

[0047] Secondly, the present invention also provides a digital supervision method for the highway and waterway construction market, the method comprising:

[0048] S1. Establish globally unique digital identities for all participants and build a dynamically linked trusted archive.

[0049] S2. Extract project information and personnel identity information from the trusted archive, and combine them with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records.

[0050] S3. Construct a digital twin that integrates highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records.

[0051] S4. Based on the progress and quality linkage analysis information, import environmental forecast data into the digital twin to simulate the future engineering process and output a simulation analysis report of the current engineering project.

[0052] S5. Construct smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

[0053] The beneficial effects of this invention are as follows:

[0054] 1. By integrating multiple modules for collaborative work, the level of intelligent project supervision is improved; by utilizing digital identity authentication and dynamic archive construction, the uniqueness and credibility of information of all participants are ensured, reducing the risk of identity fraud and data tampering; at the same time, by combining IoT devices to collect construction data in real time, violations are automatically identified and structured construction records are output, which greatly improves the real-time performance and accuracy of supervision; in addition, through project twin tracking and risk simulation modules, the linkage analysis of progress and quality, as well as the prediction of future risks, are realized, providing managers with a scientific basis for decision-making, optimizing resource allocation as a whole, and reducing the probability of project delays and quality accidents.

[0055] 2. By using multi-source data fusion technology, heterogeneous information such as project information, personnel identity, and IoT monitoring data are standardized, eliminating the information silo problem in traditional supervision. This enables the continuous flow of data throughout the entire lifecycle, from market access to construction completion, enhancing the comprehensiveness and transparency of supervision, thereby accelerating the problem rectification and feedback cycle.

[0056] 3. The construction of digital twins and the import of environmental data can simulate potential risks under different engineering projects, output quantitative analysis reports and optimization suggestions, which helps to reduce the impact of emergencies and thus build a fairer and more efficient construction market ecosystem. Attached Figure Description

[0057] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0058] Figure 1This is a system principle block diagram of a digital supervision system for the highway and waterway construction market according to an embodiment of the present invention;

[0059] Figure 2 This is a flowchart of the enterprise access registration application process in a digital supervision system for the highway and waterway construction market according to an embodiment of the present invention;

[0060] Figure 3 This is a flowchart illustrating the process of changing personnel information in a digital monitoring system for the highway and waterway construction market, according to an embodiment of the present invention.

[0061] Figure 4 This is a flowchart illustrating the process of changing expert information in a digital monitoring system for the highway and waterway construction market, according to an embodiment of the present invention.

[0062] Figure 5 This is the overall process of the construction market certification and archiving module in a digital supervision system for the highway and waterway construction market according to an embodiment of the present invention;

[0063] Figure 6 This is a flowchart of a digital supervision method for the highway and waterway construction market according to an embodiment of the present invention.

[0064] The icons in the attached diagram are labeled as follows: 1. Market Authentication and Archiving Module; 2. IoT Sensing and Inspection Module; 3. Engineering Twin Tracking Module; 4. Risk Simulation and Decision-Making Module; 5. Credit Assessment and Payment Module; 6. Visualization and Decision-Making Module; 7. Open Extension and Integration Module; 8. Security Management and Assurance Module. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0066] Please see Figure 1 This paper presents a digital supervision system for the highway and waterway construction market, which includes: a market certification and archiving module 1, an IoT sensing and inspection module 2, an engineering twin tracking module 3, a risk simulation decision-making module 4, a credit assessment and payment module 5, a visualization decision-making module 6, an open extension integration module 7, and a safety management and security module 8.

[0067] Market Authentication and Archiving Module 1 is used to establish globally unique digital identities for all participants and to build a dynamically linked trusted archive.

[0068] Specifically, the market certification and archiving module is a core component of the digital supervision system for the highway and waterway construction market. It aims to establish globally unique digital identities for all participants (including enterprises, personnel, and experts) and build a dynamically linked, trusted archive. Through standardized processes and information technology, the market certification and archiving module achieves full lifecycle management from application for access and information approval to dynamic updates, ensuring data authenticity, traceability, and real-time performance.

[0069] The following section provides a detailed explanation of the market certification archiving module's functions, in conjunction with the accompanying diagrams and specific content.

[0070] I. Enterprise Database Management: Establishing enterprise digital identity and qualification files

[0071] Enterprise database management is the foundation of market certification archiving. Through an online application and approval process, a unique digital identity is assigned to each enterprise. Enterprises must submit basic information, such as company name, unified social credit code, and legal representative information, on a designated page, along with supporting qualification documents, such as business licenses and qualification certificates. After submission, backend administrators verify the information to ensure accuracy. Upon approval, the enterprise is added to the database and a unique digital identifier is generated. When enterprise information changes, such as adjustments to qualification levels, resubmission for approval is required to ensure dynamic updates to the database. This process, through strict process control, avoids the risk of data tampering and creates a reliable enterprise qualification file.

[0072] The enterprise access approval process clearly defines the responsibilities of each step through flowcharts, such as... Figure 2 As shown.

[0073] The enterprise database also supports information statistics and query functions. Administrators can view basic enterprise information, qualification details, and project participation records, providing data support for enterprise performance and credit evaluation.

[0074] II. Personnel Database Management: Enabling Digital Identity and Tracking of Personnel

[0075] The personnel database management focuses on establishing digital identities for employees, binding personnel information through unique identifiers such as ID card numbers to ensure globally unique personnel identities. Enterprises can manage their employee information: when adding new personnel, they submit attachments such as name, ID card number, job title, and employment contract; when making changes, the system searches the existing personnel database for correlation. After approval, the personnel are added to the personnel database, and their work trajectory, performance records, and performance evaluation information are recorded. The system supports multi-dimensional queries, such as detailed personnel information, project participation history, and real-time location trajectory, forming a dynamic personnel profile database.

[0076] For example, the process of changing personnel information is standardized through flowcharts, such as... Figure 3 As shown.

[0077] The personnel database is linked to the enterprise database to ensure that personnel identities are linked to enterprise qualifications, providing a foundation for subsequent project performance. The system regularly checks the association of personnel with multiple enterprises to prevent identity conflicts and improve the reliability of the database.

[0078] III. Expert Database and Project Database Management: Expanding Digital Identity Application Scenarios

[0079] The expert database manages information on industry experts through registration or import. After experts submit basic information, such as their professional field and title, the system reviews and assigns a unique digital identity. The expert database supports random selection, automatically selecting experts based on project needs and recording their review and performance to create expert credit profiles. The project database manages all basic project information, such as name, investment amount, and bid section. Administrators create projects and link them to the command center account. The project database provides functions such as project status inquiry and performance display, ensuring dynamic linkage between project information and participant identities. Overall, the expert and project databases are interconnected with the enterprise and personnel databases, constructing a digital identity system covering all participants.

[0080] For example, the process for changing expert information is standardized through flowcharts, such as... Figure 4 As shown.

[0081] IV. Performance Information Management: Achieving Dynamic Linkage of the Archives

[0082] Contract performance information management is a core component of the trusted archive, forming a network of performance relationships by linking projects, companies, and personnel. After the project command center configures the bidding sections and winning companies, the companies assign personnel to specific positions, and the system automatically records the performance status. Performance data is updated in real time and linked to supervision and inspection functions: for example, issues discovered during inspections are pushed to the project command center, and performance scores are updated after rectification feedback. This dynamic linkage mechanism ensures the real-time nature and accuracy of the archive, providing a data foundation for credit assessment.

[0083] The overall process of the construction market certification and archiving module demonstrates a closed loop from identity establishment to performance supervision, such as... Figure 5 As shown.

[0084] The market certification and archiving module establishes a globally unique digital identity for all participants in the highway and waterway construction market through the collaborative management of enterprise, personnel, expert, and project databases. Relying on performance information management and dynamic update mechanisms, it constructs a dynamically linked and trusted archive. This module not only improves data accuracy and regulatory efficiency but also ensures system traceability and tamper-proofing through the standardized processes shown in the attached diagram, providing a solid foundation for the digital regulatory system.

[0085] Based on the overall platform architecture, the market certification and archiving module data is interconnected through the cloud platform, supporting full lifecycle supervision, which is in line with the development concept of "coordination, innovation, openness, and integration".

[0086] The IoT sensing and inspection module 2 is used to extract project information and personnel identity information from the trusted archive, and combine it with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records.

[0087] In the description of this invention, the IoT sensing and inspection module 2 includes: a multi-source data fusion module, a heterogeneous device acquisition module, a violation identification and analysis module, and a record generation and distribution module.

[0088] Specifically, the IoT Sensing and Inspection Module 2 is a core component of the digital supervision system for the highway and waterway construction market, responsible for on-site supervision. By connecting the virtual digital system with the physical construction site, it enables real-time and intelligent supervision and inspection of the entire project process. As a highly integrated system, the IoT Sensing and Inspection Module 2 operates following a coherent process from data collection and intelligent analysis to decision execution.

[0089] The IoT Sensing and Inspection Module 2, through multi-source data fusion, extracts, cleans, and correlates basic project information, personnel identity data, and contract performance relationships scattered across different systems, constructing a standardized qualification data system and establishing a unified and reliable data foundation for the entire inspection process. Simultaneously, leveraging heterogeneous device acquisition capabilities, it widely connects to various IoT sensing devices deployed on construction sites, such as surveillance cameras, environmental sensors, and positioning devices. It automatically identifies devices from different manufacturers and using different protocols, and adds precise three-dimensional spatial coordinates and timestamps to the raw data collected by these devices. This ensures that each piece of data is accurately understood within a specific spatiotemporal context and associated with specific project, personnel, or equipment files. For example, in practical applications in the power industry, edge IoT agent devices can achieve the collection and standardized transmission of data from multiple terminals.

[0090] With complete data, the core of the IoT Sensing and Inspection Module 2—the violation identification and analysis function—begins to function. It deeply integrates static qualification data from a trusted archive with dynamic IoT monitoring data, and makes intelligent judgments based on a pre-set analysis model targeting common violation scenarios in highway and waterway engineering. Combining knowledge graphs and artificial intelligence technologies, it verifies the legality of real-time operations by comparing behavioral patterns with a rule base. It also assesses the confidence level of initially identified abnormal events, outputting an event record set with a confidence score.

[0091] Ultimately, the system's record generation and distribution functions integrate verified violation records with other normal construction activity records, packaging them into structured construction records that can serve as electronic archives. Based on the event type and urgency, the system automatically selects the most appropriate notification channel, such as platform instant alerts, SMS, or app push notifications, and accurately pushes the information to relevant project managers, supervisors, or regulatory units via encrypted channels. This forms a complete business loop from perception and analysis to alert handling, enabling efficient violation detection and handling, and providing solid data support for project transparency, traceability, and subsequent credit evaluation through full-process traceability and structured records.

[0092] The multi-source data fusion module is used to establish a data mapping model, identify and associate the relationships between projects, personnel and equipment from different data sources, and output standardized qualification data to extract basic project information, personnel identity data and contract performance relationships from a trusted archive.

[0093] The heterogeneous device acquisition module is used to identify the types of IoT devices connected, add three-dimensional coordinates and timestamps to the IoT monitoring data collected by IoT devices, and establish spatial association with qualification data.

[0094] The violation identification and analysis module is used to integrate qualification data and IoT monitoring data, establish a violation identification and analysis model based on the preset violation patterns of highway and waterway engineering, verify the legality of personnel and equipment, and output an event record set containing confidence scores by combining confidence verification and filtering.

[0095] In the description of this invention, qualification data and IoT monitoring data are integrated to establish a violation identification and analysis model based on preset violation patterns in highway and waterway engineering. The model verifies the legality of personnel and equipment, and, combined with confidence level verification and filtering, outputs a set of event records containing confidence scores, including:

[0096] Step S11: Synchronize qualification data and IoT monitoring data using engineering project semantic alignment technology to obtain a real-time data stream that is uniformly aligned in terms of time, space and semantics. Then, analyze the behavior patterns of personnel and equipment in the current engineering project through dynamic permission verification and output the permission verification results.

[0097] Specifically, step S11 integrates static qualification data (including enterprise qualifications, personnel certificates, equipment registration information, etc.) from the market certification archiving module with dynamic monitoring data (including personnel GPS location, equipment operating status, video stream) collected by IoT devices through engineering project semantic alignment technology. This not only involves simple matching on timestamps and physical spatial coordinates, but also establishes associations at the business semantic level. For example, it binds a device with a specific number to its real-time transmitted operating parameters and other monitoring data, and confirms that it is associated with the engineering section currently under construction, thereby forming a real-time data stream that is unified and computable in time, space, and business context.

[0098] Building upon this foundation, the system performs dynamic permission verification by continuously analyzing the behavioral patterns of personnel or equipment within a specific project context. For example, the system verifies whether a certified welder is operating a registered welding machine within their authorized work area and whether their current time falls within a reasonable construction period. Dynamic permission verification is an extension of the Attribute-Based Access Control (ABAC) concept, integrating multiple dimensions such as user identity, device status, and environmental context for real-time analysis.

[0099] Step S12: Based on the project map-driven violation identification and analysis model, the permission verification results and IoT monitoring data are integrated. By reasoning the pre-set highway and waterway violation patterns through the map, the violation events of the current project are identified, and an asynchronous event list is output.

[0100] Specifically, the system matches entities in the real-time data stream generated in step S11 with nodes in a pre-built knowledge graph of highway and waterway engineering projects to accurately pinpoint the specific project context of the current activity. The knowledge graph embeds industry standards, construction process standards, and common violation patterns, such as unlicensed work, exceeding permitted scope of work, equipment misuse, and lack of safety measures. Subsequently, the system utilizes advanced algorithms such as graph neural networks to traverse the associated paths originating from the current project node, calculating the matching degree between the behavioral sequences captured in the real-time data stream and the pre-defined violation patterns in the knowledge graph. This allows the system to understand the context and temporal relationships of the violations. For example, the knowledge graph defines a violation pattern: "During the excavation phase, unauthorized personnel enter a hazardous area." The system analyzes in real time whether the current project phase is "excavation" (from project progress data), whether the identified personnel are "authorized," and whether their location is within a "hazardous area." When all these conditions are met simultaneously, a potential violation event is identified.

[0101] Finally, the system performs a temporal logic judgment on the initially identified suspicious events. By tracing back the scene state changes of preceding and subsequent frames, it analyzes the continuity of the entire construction operation cycle to eliminate instantaneous and accidental anomalies, ensuring the integrity and authenticity of the violation events in terms of temporal logic. All verified violation events are encapsulated into structured data objects, typically containing information such as event type, involved parties, time, location, and evidence chain, and are summarized into an asynchronous event list to prepare for subsequent confidence calibration and filtering.

[0102] In the description of this invention, a violation identification and analysis model driven by an engineering project graph is used to integrate permission verification results and IoT monitoring data. Through graph inference of preset highway and waterway violation patterns, abnormal behavior of the current engineering project is identified, and an asynchronous event list is output, including:

[0103] Step S121: Match the entities in the current real-time data stream with the nodes in the knowledge graph of highway and waterway engineering projects to identify the engineering project where the personnel and equipment are currently located.

[0104] Step S122: Based on the current engineering project, activate the preset monitoring targets in the knowledge graph, and use the cross-modal attention mechanism to extract the device behavior features and environmental operation parameters from the database, align them with the permission verification results in a unified vector space, and obtain a multimodal feature vector.

[0105] Step S123: Traverse the violation pattern path starting from the current project node, use graph neural network to calculate the matching degree between real-time data and the pre-set violation pattern in the graph, and obtain the suspected event set.

[0106] Step S124: Based on the suspected event set, analyze the temporal logic of the entire construction operation cycle by tracing back the scene state changes of preceding and subsequent frames to determine the temporal logic of the suspected events, and output the violations that have been verified by the temporal logic. The violations are then encapsulated into structured data objects and merged to generate a structured list of asynchronous events.

[0107] Step S13: Based on the asynchronous event list, perform confidence calibration on the violation events in various engineering projects, calculate the probability score in combination with environmental factors, and output the event record set that meets the confidence threshold requirements.

[0108] In the description of this invention, based on an asynchronous event list, confidence levels are calibrated for asynchronous events in various engineering projects. Probability scores are calculated by combining environmental factors, and an event record set that meets the confidence threshold requirements is output, including:

[0109] Step S131: Parse and preprocess the asynchronous event list, classify the events according to different project types based on the project identifier, and generate an event dataset grouped by project.

[0110] Step S132: Based on the timestamp and spatial coordinates of each event, dynamically extract the corresponding environmental factor data, and normalize the environmental factor data to form an environmental factor feature vector.

[0111] Step S133: Establish a prior probability distribution based on historical data, calculate the posterior probability by combining real-time environmental factors, configure different weight coefficients for different engineering projects, and calculate the post-calibration confidence of each event using Bayes' formula to generate a time probability matrix containing the post-calibration confidence score.

[0112] Step S134: Dynamically adjust the confidence threshold according to the event type and project, and filter out violations below the preset threshold using a multi-level filtering strategy. Perform risk assessment and classification on the retained violations, and generate an event record set through structured encapsulation.

[0113] The record generation and distribution module is used to integrate the identification results of the event record set with other non-abnormal event sets, encapsulate them into structured engineering construction records, and automatically select push channels according to the violation type and urgency of the event record set, and push them to the designated terminal through an encrypted channel.

[0114] The Engineering Twin Tracking Module 3 is used to construct a digital twin of highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records.

[0115] Specifically, the Engineering Twin Tracking Module 3 constructs a dynamic digital mirror that runs synchronously with the physical project and interacts with it, enabling in-depth perception, precise mapping, and intelligent insight into the entire process of the project. The Engineering Twin Tracking Module 3 is not a static 3D model display, but a complex system driven by real-time data. Its operation encompasses the digital mapping of all elements from physical entities to virtual space, real-time status perception and tracking, and the fusion and analysis of multi-dimensional data.

[0116] The Engineering Twin Tracking Module 3, based on Building Information Modeling (BIM) and high-precision 3D modeling technology, integrates real-time data streams from IoT devices to construct a 3D digital twin corresponding to the physical engineering project. This digital twin achieves comprehensive dynamic mapping of the project's physical state, personnel and equipment locations, construction progress, and even environmental conditions. When the IoT sensing and inspection module identifies violations or quality anomalies, the Engineering Twin Tracking Module 3 can immediately and automatically adjust texture attributes at the corresponding location in the digital twin by changing model color, highlighting, or adding warning labels, thereby providing a visual and intuitive early warning of abnormal conditions.

[0117] In terms of status tracking, the engineering twin tracking module 3 employs a multi-object tracking algorithm to continuously track the trajectories of moving entities such as personnel and machinery at the construction site. By deeply correlating the real-time acquired trajectory information with violations in the engineering construction records and leveraging analytical tools such as graph neural networks, the module can accurately assess the cascading impact of individual violations on subsequent construction processes, thereby quantifying their potential risks to the overall schedule. For example, it can simulate the delay effect on the critical path schedule caused by the shutdown of specific equipment due to violations.

[0118] More importantly, the core function of the Project Twin Tracking Module 3 lies in realizing the safety linkage analysis of schedule and quality. It effectively integrates traditional schedule data, real-time quality inspection indicators, and structured construction records. By constructing a schedule-quality coupling analysis model, it can deeply reveal the quality risks that schedule pressure may cause, or the specific impact of existing quality problems on subsequent schedules. Through deep correlation analysis, it can output crucial schedule-quality linkage analysis information, helping project managers understand the internal operating logic of the project, optimize decisions, and thus promote the smooth implementation of the project while ensuring project quality and safety.

[0119] In the description of this invention, the engineering twin tracking module 3 includes: a twin construction module, a status tracking and detection module, and a progress and quality linkage module.

[0120] The twin construction module is used to build a three-dimensional digital twin based on real-time data streams and building information models, enabling dynamic visualization and entity mapping of engineering projects. When a violation event is detected in the engineering construction record, the corresponding texture attributes in the digital twin are automatically adjusted.

[0121] The status tracking and detection module is used to track the real-time status of personnel and equipment in the digital twin, and integrate violations in the engineering construction records to analyze the potential impact of violations on the construction progress.

[0122] In the description of this invention, tracking the real-time status of personnel and equipment in a digital twin, and integrating violations from engineering construction records to analyze the potential impact of violations on construction progress includes:

[0123] Step S211: Use a multi-target tracking algorithm to continuously track the trajectory of personnel and equipment in the digital twin, and combine video stream data to perform behavior pattern recognition to obtain the group operation mode.

[0124] Specifically, the core objective of step S211 is to achieve continuous status monitoring of personnel and equipment within the digital twin and identify their behavioral patterns, laying a data foundation for subsequent correlation analysis. Data is collected in real time through various sensing devices deployed at the physical construction site. For example, equipping each construction worker with a safety helmet containing a built-in UWB chip and installing sensors on critical equipment (such as tower cranes and pump trucks) allows for real-time acquisition of data such as their three-dimensional coordinates, movement speed, and orientation, accurate to the centimeter level. Simultaneously, the video surveillance system provides continuous visual stream data. Utilizing multi-target tracking algorithms (such as SORT or Deep SORT algorithms) to process these real-time data streams enables the correlation and fusion of data from different sensors regarding the same person or equipment. Even when the line of sight is briefly obstructed or targets intersect, tracking continuity is maintained, thereby generating a smooth and continuous motion trajectory within the digital twin.

[0125] Based on continuous trajectories, behavioral pattern recognition is performed using video stream data. Computer vision techniques, such as object detection and behavior classification models, are employed to analyze the behavioral information contained within the trajectories. For example, the system can identify patterns such as personnel gathering in specific areas and equipment remaining stationary for extended periods. Through cluster analysis of large amounts of historical trajectory and behavioral data, typical group operation patterns can be further extracted, such as the collaborative paths of various trades during concrete pouring and the regular routes of material transport vehicles. Finally, these dynamic trajectories and identified patterns are visualized in a digital twin, providing managers with intuitive on-site situational awareness.

[0126] Step S212: Deeply correlate the violations in the construction records with the real-time behavior of the entities, establish an event behavior association rule base, calculate the correlation between events and behaviors through graph neural networks, and identify the scope of the impact of violations on the construction process.

[0127] Specifically, step S212 aims to correlate static construction records (such as violation reports) with the dynamic, real-time behavior of entities in the digital twin, quantifying the impact of violations on the construction process. This process can be divided into three stages:

[0128] The first phase involves establishing a rule base for event-behavior association: This involves structuring violations in historical engineering records, clearly defining each type of event, its associated entities (people, equipment), and typical behavioral manifestations. For example, the event "working at height without a safety belt" is associated with the entity "workers at height," and a typical behavior might be moving at the edge of the platform without the system detecting any safety rope tension data. These rules form the knowledge base for association analysis.

[0129] The second stage involves real-time correlation and graph neural network (GNN) computation: When the real-time tracking system identifies a potential violation, it immediately triggers correlation analysis. The system constructs a temporary relationship graph from various data points in the current scenario. Subsequently, a GNN is used to process this graph. The advantage of GNN lies in its ability to aggregate information about nodes (the current actor) and their neighboring nodes (other people and devices in the vicinity), as well as edges (interrelationships), in the graph. Through a message passing mechanism, it calculates the correlation between the current behavior and various violation events in the rule base. The correlation is a quantifiable value representing the probability that the behavior will lead to a certain type of violation event.

[0130] The third stage is impact scope identification: Based on the high correlation calculated by GNN, the system can not only confirm the occurrence of violations but also analyze their impact scope. For example, the "overloading" behavior of a tower crane, through GNN analysis, may identify its chain effects on the safety of personnel in the work area below and spatial conflicts between adjacent work surfaces. This achieves a sublimation from isolated event identification to systemic risk insight.

[0131] Step S213: Based on the results of the event behavior correlation analysis and combined with the schedule data, assess the potential impact of the violation event on the construction progress, and conduct multi-scenario impact simulation by quantifying the delay time of a single violation event, and output the risk quantification index of the violation event.

[0132] Specifically, the first step is to conduct an impact assessment and quantify the delay time. The system maps the identified violations and their scope of impact to the project schedule. For example, if the violation is "a material transport vehicle on the critical path malfunctions due to improper operation," the system will assess the time required to handle the malfunction and the delays caused to subsequent processes based on preset rules and historical data, thereby quantifying the potential delay time of this single violation.

[0133] Furthermore, it should be noted that the impact assessment and delay time quantification described in this invention also include a resource scheduling optimization unit, which improves resource utilization by optimizing the scheduling algorithm. Based on the simulated scenario, the construction sequence is automatically adjusted. For example, when equipment failure affects the critical path, the delay time is reduced to within a preset delay threshold after algorithm optimization, while providing real-time dynamic updates. For example, "In the equipment failure scenario, the critical path time delay is approximately 5 hours," and subsequent optimized paths are provided. The resource scheduling optimization unit specifically includes:

[0134] In the twin building module, performance thresholds are set for each device in the project, and the operating parameters of each device are monitored in real time. Anomaly detection algorithms automatically identify faults in each device. Positioning technology is used to monitor the readiness status of each device; if a device deviates from its designated location coordinates, it is marked as an absent device.

[0135] Using digital twins and schedule models, the impact of equipment failures or absences on project tasks is calculated, specifically including:

[0136] Each device is treated as a node. Project management methods are used to identify the task paths of each device within the project and treat them as edges. The edge relationships of each path are extracted. It should be noted that, in this invention, edge relationships include time dependencies, resource dependencies, logical dependencies, and causal relationships. The earliest power-on time, latest power-off time, and total float time of each device are obtained. Based on the earliest power-on time and latest power-off time, the duration of each task path is calculated. The degree of prerequisite dependencies for each task path is determined based on the edge relationships of each device. The resource constraints of each task path are statistically analyzed. The duration, degree of prerequisite dependencies, and resource constraints of each task path are used as key evaluation parameters for each task path.

[0137] The key evaluation parameters for each task path are analyzed and processed, specifically including:

[0138] Methods for quantifying the degree of pre-dependency:

[0139] The number of other task paths that a given task path depends on before it begins execution is counted, and these task paths are recorded as dependent task paths. Simultaneously, the duration of each dependent task path is obtained, and the duration of each dependent task path is input into a pre-stored mapping range of duration-predependency weights in the database for mapping and matching to obtain the predependency weight of each dependent task path. In this embodiment of the invention, task path A depends on task paths B, C, and D. The predependency weights of task paths B, C, and D are obtained, and the sum of these predependency weights is multiplied by the total number of dependent task paths to obtain the predependency degree of task path A.

[0140] Resource constraint quantification methods:

[0141] The resource requirements of each task path are calculated. Since the types of resources required by each task path are different, the resource requirements are normalized. By using a preset unit conversion rule, the various types of resources used in the project are converted into common units, which are defined as resource consumption. The resource requirements of each task path are then used to obtain the resource consumption of each task path.

[0142] Obtain the mean of the critical evaluation parameters for all task paths in the project, and count the task paths for which all critical evaluation parameters exceed the corresponding mean, and record them as critical task paths.

[0143] A delay prediction model is established based on historical data and machine learning algorithms. Inputting historical failure data, the model analyzes failure data from historical engineering projects to predict the delay impact of similar events on project schedule. Based on equipment failure or absence, the delay time of each task is dynamically tracked; the delay time calculation expression is as follows:

[0144] ;

[0145] in, The delay time for task path Y. This represents the earliest actual start time for task path Y. The earliest originally scheduled start time for task path Y. The delay time for the X task path.

[0146] If the delay in task path Y affects the start time of task path Z, then continue propagating the delay and calculating... And so on, to obtain the predicted delay time for each task path.

[0147] Any delay in a task on the critical task path will directly affect the overall project duration. Therefore, it is necessary to extract the predicted delay time of the critical task path and dynamically adjust resource scheduling based on the predicted delay time to reduce the impact of delay time. When a delay is predicted for a task on a critical task path, the progress and delay status of each task can be dynamically tracked using digital twins and schedule models, and the impact of delay can be pushed out in real time.

[0148] Combining delay prediction, resource scheduling optimization, and digital twins can effectively improve the execution efficiency and flexibility of engineering projects. Through accurate delay prediction and automated scheduling optimization, task progress and resource allocation can be monitored in real time, reducing the impact of equipment failures and absences on projects and ensuring timely and high-quality completion. Furthermore, by improving resource utilization, reducing delay risks, and optimizing critical path management, engineering projects can ultimately achieve more efficient implementation and better economic benefits.

[0149] Further, multi-scenario impact simulations are conducted. To more comprehensively assess risks, the system performs multi-scenario simulations. It doesn't just base its simulations on the outcome of a single event, but considers multiple possibilities. For example, it simulates whether the delay time will change under different resource allocation plans; or, if multiple related violations occur simultaneously, what impact will their cumulative effect or chain reaction have on the schedule? This simulation helps managers anticipate the worst, best, and most likely scenarios.

[0150] The final output is a quantitative risk indicator. Based on the above analysis, the system will generate a series of intuitive quantitative risk indicators to support decision-making. These indicators may include:

[0151] 1. Delay probability distribution: Predict the probability of different durations of progress delay (e.g., 0 hours, 2-4 hours, more than 4 hours).

[0152] 2. Critical Path Impact Index: Measures the degree of impact of the event on the critical path of the project.

[0153] 3. Risk Level: Based on the combined delay time and probability of occurrence, the event risk is divided into "high", "medium" and "low" levels.

[0154] The schedule and quality linkage module is used to integrate schedule data, quality inspection indicators, and engineering construction records to perform multi-dimensional linkage analysis and output schedule and quality linkage analysis information.

[0155] In the description of this invention, progress data, quality inspection indicators, and engineering construction records are integrated to perform multi-dimensional linkage analysis, and the output progress-quality linkage analysis information includes:

[0156] Step S221: Obtain the progress data and quality inspection indicators of the current project, align them with the project construction records in time and space to form a standardized data stream, and dynamically adjust the weight coefficients of quality and progress according to the project stage to construct a progress-quality coupling analysis model and output the progress-quality correlation matrix.

[0157] Specifically, step S221, the data preparation and core calculation stage of the entire linkage analysis, aims to integrate massive amounts of data from different sources and with different structures into a computable analytical model that accurately reflects the intrinsic relationship between progress and quality. First, the system acquires three types of core data from various subsystems: progress data reflecting the actual progress of the project, quality inspection indicators characterizing the physical quality of the project, and project construction records documenting historical and current activities. These data are often asynchronous in terms of timestamps and spatial coordinates; therefore, the system needs to unify the data to the same spatiotemporal benchmark through spatiotemporal alignment technology. For example, the compaction data detected on a specific road section on a certain day is precisely matched with the paving progress of that section on that day and whether there are any records of violations, forming a standardized data stream that is completely synchronized in time and space.

[0158] Building upon this foundation, a schedule-quality coupling analysis model is constructed. The key is recognizing that the importance of schedule and quality is not static but dynamically evolves with each stage of the project. For example, during the foundation construction phase, the weighting coefficient for quality may be set extremely high because potential quality issues at this stage can have a fatal impact on all subsequent work. Conversely, in non-critical path processes such as interior decoration, the weighting of schedule may be appropriately increased. The system dynamically assigns appropriate weighting coefficients to quality and schedule indicators for different project stages based on a pre-defined rule base or machine learning algorithm. Finally, the model outputs a schedule-quality correlation matrix that intuitively demonstrates the relationship between schedule and quality through multivariate calculations. This matrix can quantitatively reveal deeper patterns, such as "when the road paving speed exceeds a certain threshold, the smoothness pass rate will show a downward trend," providing managers with a basis for decision-making.

[0159] Step S222: Map the violations in the project construction records to the schedule and quality correlation matrix, use graph neural networks to analyze the causal relationship of the event chain, analyze the causes of schedule delays or quality problems, and obtain schedule and quality linkage analysis information.

[0160] Specifically, the specific violations in the construction records are mapped to the progress-quality correlation matrix generated in step S221, so that each violation is no longer an isolated record, but is accurately located to the specific progress link and quality dimension affected by it in the matrix.

[0161] The system utilizes graph neural network (GNN) artificial intelligence technology for analysis. GNN excels at handling graph-structured data with complex relationships. In this scenario, the system constructs a complex "project knowledge graph" from various tasks, resources, and violations within the project. Once a violation is mapped to the graph, the GNN can simulate and infer the potential "ripple effects" of this event along the connection paths within the graph, analyzing the causal relationships between events. For example, it infers that "incompetent personnel qualifications" (violation event A) led to "substandard welding processes" (quality event B), which in turn triggered "component rework" (schedule event C), ultimately causing a chain reaction of "project delays." Through this analysis, the system can accurately pinpoint the root cause of schedule delays or quality problems, rather than superficial phenomena.

[0162] All analysis results are aggregated to generate schedule-quality linkage analysis information, which not only indicates the current coupling status of schedule and quality, such as "quality is better than schedule" or "severe schedule delay and high quality risk", but also clearly reveals the key event chain that led to this status, providing project managers with direct and accurate decision-making intervention targets, thereby achieving an upgrade from passive response to proactive early warning and precise intervention.

[0163] Risk simulation decision module 4 is used to simulate future engineering processes by importing environmental forecast data into the digital twin based on schedule and quality linkage analysis information, and output a simulation analysis report of the current engineering project.

[0164] Specifically, the risk simulation decision-making module 4 first integrates progress and quality linkage analysis information reflecting the current status of the engineering project from the engineering twin tracking module, thereby constructing a dynamic virtual model in the digital space that is completely synchronized with the physical entity of the project. Next, it imports external environmental forecast data, such as weather forecasts, hydrological data, and geological change trends for a future period. Through simulation models built using theories such as computational fluid dynamics and computational mechanics, the module can simulate the potential impact of these environmental factors on future engineering processes on the digital twin.

[0165] The risk simulation decision-making module 4's operation process includes scenario setting, parameter adjustment, simulation and result generation. It can set various scenarios, such as earthwork excavation under heavy rain conditions or bridge hoisting under specific wind conditions, and observe the simulation results under different decision-making schemes by adjusting parameters. Finally, the module generates a structured simulation analysis report that quantitatively assesses the impact of various risks on project schedule, cost, quality, and safety.

[0166] The core function of Risk Simulation Decision Module 4 is to transform post-event response into pre-event early warning, providing forward-looking decision support for project management. It helps managers preview the future of the project, intuitively assess the feasibility and risks of different construction schemes in complex environments, thereby optimizing construction organization and resource allocation. Quantitative analysis of potential risks helps develop more targeted contingency plans, significantly improving the project's ability to cope with uncertainty, effectively preventing schedule delays, cost overruns, and quality and safety accidents, and ensuring the smooth progress of the project.

[0167] The credit assessment and payment module 5 is used to build smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

[0168] Specifically, the credit assessment and payment module 5 constructs smart contracts that are linked to industry credit and performance. Smart contracts are pre-written, automatically executed computer program codes whose execution conditions are directly linked to the credit scores and contract performance of the participants.

[0169] The credit assessment and payment module 5 continuously collects data from the market certification and archiving module, including enterprise qualifications, personnel information, historical performance, violation records reported by the IoT sensing and auditing module, and performance quality assessed by the engineering twin tracking module. It then uses a credit assessment model and algorithm to perform a comprehensive credit score on the participants. When the pre-set payment conditions in the smart contract are met, such as the successful acceptance of a project milestone and the contractor's credit score meeting the requirements, the module automatically generates a payment instruction, triggering subsequent financial processes and achieving automated settlement based on credit value.

[0170] The Credit Assessment and Payment Module 5 aims to establish a market-based mechanism of "incentivizing trustworthiness and punishing dishonesty." By directly linking credit rating results with actual economic benefits (payment processes), credit becomes a "realizable asset" for participants, effectively incentivizing them to regulate their behavior, prioritize performance quality, thereby purifying the market environment and reducing transaction risks.

[0171] The Visual Decision Module 6 is used to build a configurable visualization interface through graphical interaction, and supports graph-data linkage and data drill-down functions to enable interactive interaction among personnel.

[0172] Specifically, the Visualization Decision Module 6 relies on data visualization and graphical human-computer interaction technologies. Through a configurable graphical interface, it presents complex, multi-dimensional data from the monitoring system in intuitive forms such as charts, maps, and 3D models. It supports linked chart and data functionality, meaning that when users interact with graphical elements, the associated data changes and is filtered in real time. It also provides data drill-down capabilities, allowing users to drill down from a summary chart to more detailed data levels, such as daily inspection records for a specific project. Through well-designed interaction methods, it achieves a seamless transition from macro to micro perspectives, providing managers with data tools for in-depth analysis.

[0173] The core function of the Visual Decision-Making Module 6 is to lower the decision-making threshold and improve decision-making efficiency and scientific rigor. It transforms abstract data into easily understandable visual information, helping managers at all levels quickly and comprehensively grasp the overall situation and accurately pinpoint problems. Through interactive engagement, managers can explore and simulate various types of information, much like operating a "cockpit," thereby gaining a deeper understanding of the business logic behind the data and supporting scientific and efficient command and decision-making. It acts as the "intelligent command center" of the entire digital supervision system, enhancing the transparency and collaborative efficiency of supervision.

[0174] The open extension integration module 7 is used to build connection channels for external ecosystem connections. Through standardized interfaces and security protocols, it provides application plugins and analysis tools for development, testing, and platform.

[0175] Specifically, the core technology of the Open Extension Integration Module 7 is standardized interfaces and security protocols. It is responsible for building channels connecting to the external ecosystem, typically employing mature interface technologies such as RESTful APIs and message queues, and adhering to security protocols such as OAuth and HTTPS to ensure secure data exchange. It provides a complete development and testing environment, along with corresponding software development kits, facilitating third-party developers or partners to create customized application plugins or integrate new data sources and services based on the platform's core capabilities.

[0176] The Open Extension Integration Module 7 ensures the system's openness, scalability, and sustainable evolution capabilities, preventing it from becoming an "information silo." It allows for flexible integration with other business systems, regulatory platforms, or emerging IoT devices, continuously absorbing new technologies and data resources. By providing standardized integration methods, it reduces the complexity and cost of future system function expansion and technology upgrades, better adapting to changes in business needs and technological developments, thereby ensuring the entire digital regulatory ecosystem remains vibrant and continuously improving.

[0177] Security Management and Assurance Module 8 is used to deploy network security protection tools, monitor and warn of network indicators in real time, and formulate and implement data backup and disaster recovery strategies.

[0178] Specifically, Security Management and Assurance Module 8 integrates network security protection, real-time monitoring and early warning, and data backup and disaster recovery technologies. It constructs a multi-layered security defense by deploying network security tools such as firewalls, intrusion detection systems, and access control mechanisms. It monitors key indicators such as network traffic, user behavior, and system logs in real time, immediately triggering an early warning mechanism upon detecting abnormal activity or potential threats. Simultaneously, it formulates and automatically executes strict data backup strategies, regularly backing up core business data and system configurations, and establishing comprehensive disaster recovery plans to ensure rapid system recovery and minimize losses in the event of unforeseen circumstances.

[0179] Please see Figure 6 It also provides a digital regulatory approach to the highway and waterway construction market, which includes:

[0180] S1. Establish globally unique digital identities for all participants and build a dynamically linked, trusted archive.

[0181] S2. Extract project information and personnel identity information from the trusted archive, and combine them with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records.

[0182] S3. Construct a digital twin that integrates highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records.

[0183] S4. Based on the progress and quality linkage analysis information, import environmental forecast data into the digital twin to simulate the future engineering process and output a simulation analysis report of the current engineering project.

[0184] S5. Construct smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

[0185] In summary, by leveraging the technical solutions described above, the intelligent level of engineering supervision is enhanced through the integration of multiple modules working collaboratively. Digital identity authentication and dynamic archive construction ensure the uniqueness and credibility of information from all participants, reducing the risks of identity fraud and data tampering. Simultaneously, real-time collection of construction data via IoT devices automatically identifies violations and outputs structured construction records, significantly improving the real-time nature and accuracy of supervision. Furthermore, through engineering twin tracking and risk simulation modules, the linkage analysis of progress and quality, as well as the prediction of future risks, are realized, providing managers with a scientific basis for decision-making, optimizing resource allocation overall, and reducing the probability of project delays and quality accidents. Through multi-source data fusion technology, heterogeneous information such as project information, personnel identities, and IoT monitoring data is standardized, eliminating the information silos in traditional supervision. This allows for the continuous flow of data throughout the entire lifecycle, from market access to construction completion, enhancing the comprehensiveness and transparency of supervision, thereby accelerating the problem rectification and feedback cycle. The construction of digital twins and the import of environmental data can simulate potential risks under different engineering projects, output quantitative analysis reports and optimization suggestions, which helps to reduce the impact of emergencies and thus build a fairer and more efficient construction market ecosystem.

[0186] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

Claims

1. A digital monitoring system for the highway and waterway construction market, characterized in that, The system includes: The market authentication and archiving module is used to establish globally unique digital identities for all participants and to build a dynamically linked trusted archive. The IoT sensing and inspection module is used to extract project information and personnel identity information from the trusted archive, and combine it with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records. The engineering twin tracking module is used to build a digital twin of highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records. The risk simulation decision-making module is used to simulate future engineering processes by importing environmental forecast data into the digital twin based on schedule and quality linkage analysis information, and output a simulation analysis report of the current engineering project. The credit assessment and payment module is used to build smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

2. The digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The digital oversight system also includes: The visualization decision-making module is used to build a configurable visualization interface through graphical interaction, and supports the linkage of graphs and data and data drill-down functions to realize human interaction. Open extension integration modules are used to build connection channels for external ecosystems. Through standardized interfaces and security protocols, they provide application plugins and analysis tools for development, testing, and the platform. The security management and protection module is used to deploy network security protection tools, monitor and issue early warnings for network indicators in real time, and formulate and implement data backup and disaster recovery strategies.

3. The digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The IoT sensing and auditing module includes: The multi-source data fusion module is used to establish a data mapping model, identify and associate the relationships between projects, personnel and equipment from different data sources, and output standardized qualification data to extract basic project information, personnel identity data and performance relationships from the trusted archive. The heterogeneous device acquisition module is used to identify the types of IoT devices connected, add three-dimensional coordinates and timestamps to the IoT monitoring data collected by IoT devices, and establish spatial association with qualification data; The violation identification and analysis module is used to integrate qualification data and IoT monitoring data, establish a violation identification and analysis model based on the preset violation patterns of highway and waterway engineering, verify the legality of personnel and equipment, and output an event record set containing confidence scores by combining confidence verification and filtering. The record generation and distribution module is used to integrate the identification results of the event record set with other non-abnormal event sets, encapsulate them into structured engineering construction records, and automatically select push channels according to the violation type and urgency of the event record set, and push them to the designated terminal through an encrypted channel.

4. The digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The fusion of qualification data and IoT monitoring data establishes a violation identification and analysis model based on preset violation patterns in highway and waterway engineering, verifies the legality of personnel and equipment, and, combined with confidence level verification and filtering, outputs a set of event records including confidence level scores, including: By using semantic alignment technology for engineering projects, qualification data and IoT monitoring data are synchronized to obtain a real-time data stream that is uniformly aligned in terms of time, space and semantics. Dynamic permission verification is used to analyze the behavior patterns of personnel and equipment in the current engineering project and output the permission verification results. Based on the project graph-driven violation identification and analysis model, the system integrates permission verification results and IoT monitoring data. By reasoning the pre-defined highway and waterway violation patterns through the graph, it identifies violation events in the current project and outputs an asynchronous event list. Based on an asynchronous event list, confidence levels are calibrated for violations in various engineering projects. Probability scores are calculated by combining environmental factors, and an event record set that meets the confidence threshold requirements is output.

5. A digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The violation identification and analysis model based on the project graph-driven approach integrates permission verification results and IoT monitoring data. Through graph inference of pre-defined highway and waterway violation patterns, it identifies abnormal behaviors in the current project and outputs a list of asynchronous events, including: Match entities in the current real-time data stream with nodes in the knowledge graph of highway and waterway engineering projects to identify the engineering project where personnel and equipment are currently located. Based on the current engineering project, the pre-set monitoring targets in the knowledge graph are activated, and the cross-modal attention mechanism is used to extract the device behavior features and environmental operation parameters from the database, and align them with the permission verification results in a unified vector space to obtain a multimodal feature vector. Traverse the violation pattern paths starting from the current project node, use graph neural networks to calculate the matching degree between real-time data and the pre-defined violation patterns in the graph, and obtain the suspected event set; Based on the suspected event set, the temporal logic of the entire construction operation cycle is analyzed by tracing back the scene state changes of the preceding and subsequent frames to make temporal logic judgments on the suspected events, and output the violation events that have been verified by the temporal logic; and the violation events are encapsulated into structured data objects and merged to generate a structured asynchronous event list.

6. The digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The method involves calibrating the confidence level of asynchronous events in various engineering projects based on an asynchronous event list, calculating probability scores by combining environmental factors, and outputting a set of event records that meet the confidence threshold requirements. This set includes: The asynchronous event list is parsed and preprocessed, and the events are classified according to different project types based on the project identifier to generate an event dataset grouped by project. Based on the timestamp and spatial coordinates of each event, the corresponding environmental factor data is dynamically extracted and normalized to form an environmental factor feature vector. A prior probability distribution is established based on historical data, and a posterior probability is calculated by combining real-time environmental factors. Different weight coefficients are configured for different engineering projects, and the post-calibration confidence of each event is calculated using Bayes' formula to generate a time probability matrix containing the post-calibration confidence score. The confidence threshold is dynamically adjusted based on the event type and project. Combined with a multi-level filtering strategy, violations below the preset threshold are filtered out. The retained violations are subject to risk assessment and classification, and an event record set is generated through structured encapsulation.

7. A digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The engineering twin tracking module includes: The twin construction module is used to build a three-dimensional digital twin based on real-time data streams and building information models, realize dynamic visualization and entity mapping of engineering projects, and automatically adjust the corresponding texture attributes in the digital twin when violations are detected in the engineering construction records. The status tracking and detection module is used to track the real-time status of personnel and equipment in the digital twin, and integrate violations in the engineering construction records to analyze the potential impact of violations on the construction progress. The schedule and quality linkage module is used to integrate schedule data, quality inspection indicators, and engineering construction records to perform multi-dimensional linkage analysis and output schedule and quality linkage analysis information.

8. A digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The process of tracking the real-time status of personnel and equipment in a digital twin, and integrating violations from engineering construction records to analyze the potential impact of these violations on construction progress includes: Multi-target tracking algorithms are used to continuously track the trajectories of people and equipment in a digital twin, and behavioral pattern recognition is performed by combining video stream data to obtain the group operation mode. By deeply linking violations in the construction records with real-time entity behavior, an event behavior association rule base is established. The correlation between events and behaviors is calculated through graph neural networks to identify the scope of the impact of violations on the construction process. Based on the results of event behavior correlation analysis and combined with schedule data, the potential impact of violations on construction progress is assessed. By quantifying the delay time of individual violations, multi-scenario impact simulations are conducted to output risk quantification indicators for violations.

9. A digital monitoring system for the highway and waterway construction market according to claim 1, characterized in that, The integrated progress data, quality inspection indicators, and engineering construction records are used for multi-dimensional linkage analysis, and the output progress and quality linkage analysis information includes: The system acquires the current project's progress data and quality inspection indicators, aligns them with the project construction records in time and space to form a standardized data stream, and dynamically adjusts the weight coefficients of quality and progress according to the project stage to construct a schedule-quality coupling analysis model and output a schedule-quality correlation matrix. By mapping violations in the project construction records to the schedule and quality correlation matrix, and using graph neural networks to analyze the causal relationships of the event chain, the causes of schedule delays or quality problems are analyzed, and schedule and quality linkage analysis information is obtained.

10. A digital supervision method for the highway and waterway construction market, employing the digital supervision system for the highway and waterway construction market as described in any one of claims 1-9, characterized in that, The method includes: S1. Establish globally unique digital identities for all participants and build a dynamically linked trusted archive. S2. Extract project information and personnel identity information from the trusted archive, and combine them with IoT monitoring data collected by IoT devices to identify violations in highway and waterway engineering projects and output structured engineering construction records. S3. Construct a digital twin that integrates highway and waterway engineering projects, and output progress and quality linkage analysis information by combining engineering projects and engineering construction records. S4. Based on the progress and quality linkage analysis information, import environmental forecast data into the digital twin to simulate the future engineering process and output a simulation analysis report of the current engineering project. S5. Construct smart contracts that bind industry credit and performance behavior, generate payment instructions based on the credit scores of participants, and realize the market-based realization of credit value.

Images