Intelligent Compliance Management System and Method for Safety Standardization Information of Nonferrous Heavy Metals

Abstract

This invention discloses an intelligent compliance management system and method for non-ferrous heavy metal safety standardization information, belonging to the field of enterprise safety production information management technology. The method includes: collecting regulatory data for the non-ferrous heavy metal industry, generating and storing tagged regulatory data after AI filtering and tagging; decomposing the clauses in the tagged regulatory data into structured data fragments and automatically filling them into a system template to generate safety management system documents; associating the specific requirements in the safety management system documents with operational scenarios, and performing real-time monitoring and closed-loop processing by comparing on-site data with system requirements to generate execution monitoring data; and analyzing the compliance status based on the tagged regulatory data and the execution monitoring data, and generating revision suggestions for the safety management system documents. This invention achieves fully automated management from regulatory acquisition to system improvement, significantly improving the level of enterprise safety production standardization.

Description

Technical Field

[0001] This invention relates to the field of enterprise safety production information management technology, specifically to an intelligent compliance management system and method for non-ferrous heavy metal safety standardization information. Background Technology

[0002] Safety standardization is a core means for non-ferrous heavy metal smelting enterprises to fulfill their primary responsibilities and prevent systemic risks. The compliance, applicability, and effective implementation of laws, regulations, and safety management systems are the cornerstones of standardization. However, current management methods in this area within the industry suffer from the following problems: First, regulatory updates are severely lagging. The regulatory system for the non-ferrous and heavy metals industry is complex and frequently updated. Companies rely on manual collection of information from government websites at various levels and standards-issuing organizations, resulting in an average lag of 2-3 months. This leads to compliance risks where "new regulations have been implemented, but companies are still following the old ones." Furthermore, analyzing the correlation between new regulations and existing company policies is difficult, making it hard to accurately identify specific clauses that need revision.

[0003] Secondly, the safety management system is disorganized. Many enterprises use generic templates to formulate regulations, failing to closely integrate them with the specific processes of lead, zinc, and copper smelting (such as electrolysis, smelting, and anode mud treatment). This results in vague clauses and poor operability. The regulations suffer from unclear hierarchical structures, overlapping content, and even conflicts, leaving employees confused and far removed from the requirement of "unified and consistent regulations" in the "Standards for Safety Production Standardization Assessment of Non-ferrous Heavy Metal Smelting Enterprises."

[0004] Secondly, there is a lack of oversight in the implementation of regulations. The effective implementation of regulations relies heavily on self-discipline and post-event paper-based checks, lacking effective real-time monitoring and early warning mechanisms. For example, safety training programs are often abandoned due to production constraints, protective equipment wearing regulations are rendered ineffective, and supervision often only intervenes after an accident occurs, remaining in a reactive state.

[0005] Finally, there is the issue of fragmented management data. Regulations, systems, and implementation records are scattered across different departments, forming "data silos" that are difficult to analyze in relation to other data. This prevents companies from quantitatively assessing the effectiveness of their systems, and makes it impossible to achieve the core standardized requirement of "continuous improvement," leaving system optimization without data support.

[0006] Therefore, there is an urgent need for a solution that can deeply integrate industry characteristics and achieve intelligent management and control across the entire chain to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent compliance management system and method for non-ferrous heavy metal safety standardization information, in order to solve the problems of lagging regulatory tracking, chaotic system, lack of enforcement supervision, and fragmented management data in the existing technology. It realizes full-process automated intelligent management from regulatory acquisition to system improvement, and significantly improves the level of enterprise safety production standardization.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for intelligent compliance management of safety standardization information for non-ferrous heavy metals includes the following steps: S1. Build an industry-specific regulatory database: Collect regulatory data for the non-ferrous heavy metals industry, and generate and store tagged regulatory data after AI screening and tagging. S2. Generate safety management system: Decompose the clauses in the tagged regulatory data into structured data fragments and automatically fill them into the system template to generate a safety management system document; S3. Execution and Monitoring: Associate the specific requirements in the safety management system documents with the work scenarios, and conduct real-time monitoring and closed-loop handling by comparing on-site data with system requirements to generate execution monitoring data; S4. Evaluation and Improvement: Based on the tagged regulatory data and the execution monitoring data, analyze the compliance status and generate revision suggestions for the security management system documents.

[0009] Furthermore, the AI ​​screening described in S1 specifically involves: based on a preset keyword library for the non-ferrous heavy metals industry, calculating the relevance weight between the regulatory text and the industry through a semantic analysis model, and filtering out general clauses with weights lower than a preset threshold.

[0010] Furthermore, the clauses described in S2 are broken down into structured data fragments, specifically: by using natural language processing technology to identify entities, indicator values, operators, and responsible parties in the clauses, unstructured text is transformed into structured data units containing clear semantic roles, forming institutional gene encoding.

[0011] Furthermore, S2 also includes a system compatibility check, specifically: by comparing the text of the safety management system document with the compliance requirements in the tagged regulatory data through an algorithm, automatically identifying and prompting clause conflicts or incompatible content in cross-production process scenarios.

[0012] Furthermore, following S2, it also includes the release and training of regulations, specifically: based on the generated safety management regulations documents, the scope of target personnel is determined according to their tags and the documents are pushed for signature, and assessment questions are automatically generated. After the target personnel pass the assessment, their relevant work permissions are activated.

[0013] Furthermore, the specific requirements described in S3 are associated with the work scenario. By setting a unique identifier at the work point, the operator can scan the identifier with a mobile terminal to instantly access and confirm the list of system requirements and inspection items bound to that scenario.

[0014] Furthermore, the closed-loop processing described in S3 includes: Automatically generate a rectification notice containing evidence of violations and the relevant regulations, and push it to the responsible person's mobile terminal; The relevant work permissions of the responsible person are suspended until the responsible person uploads evidence of rectification. The evidence will be compared and verified with the relevant institutional provisions. Once the verification is successful, the closed-loop process will be completed and the operating privileges will be restored.

[0015] Furthermore, the generation of the revision suggestions described in S4 specifically includes: Based on the execution monitoring data, statistics are compiled on the violation records and their occurrence rate related to a specific clause; When the occurrence rate of violations related to a certain clause exceeds a preset threshold, the system automatically associates the corresponding requirements in the tagged regulatory data and outputs suggestions for adding, modifying, or deleting the clause.

[0016] Another objective of this invention is to provide an intelligent compliance management system for non-ferrous heavy metal safety standardization information. When executed, the system implements the aforementioned intelligent compliance management method for non-ferrous heavy metal safety standardization information, comprising: The data acquisition and processing module is used to collect regulatory data and process it through AI filtering and tagging to generate tagged regulatory data; The system generation and verification module is connected to the data acquisition and processing module. It is used to decompose the tagged regulatory data into system gene codes and fill them into the template to generate and verify the safety management system documents. The execution monitoring and closed-loop handling module is connected to the system generation and verification module, and is used to bind system requirements with work scenarios and perform real-time monitoring and closed-loop handling based on on-site data; The intelligent assessment and improvement module is connected to the data acquisition and processing module and the execution monitoring and closed-loop handling module, respectively, and is used to analyze the compliance status and generate suggestions for system revision. The policy release and training module is connected to the policy generation and verification module. It is used to push documents to target personnel, generate test questions, and manage assessments and permission activation after the policy is released.

[0017] Furthermore, the execution monitoring and closed-loop processing module is communicatively connected to at least one of the following: a gas sensor, a smart safety helmet, or a video monitoring device installed at the work site, to acquire on-site data in real time.

[0018] The intelligent compliance management system and method for non-ferrous heavy metal safety standardization information provided by this invention have the following advantages compared with the prior art: 1. This invention achieves real-time and accurate tracking of regulatory requirements, solving the problem of tracking lag. Traditional manual tracking of regulations has a lag period of 2 to 3 months. This invention, through an AI-driven dedicated database, can automatically capture, filter, and associate regulatory clauses with enterprise information, shortening the regulatory tracking response time to the hour level. This ensures that the enterprise's security management system is synchronized with the latest national regulations and standards in real time, eliminating "unknowing violations" from the source and significantly reducing compliance risks.

[0019] 2. It facilitates a deep integration of regulations and practical operations, resolving the problem of systemic chaos. Addressing the disconnect between corporate regulations and specific metal categories and production processes, this invention utilizes "regulatory gene coding" technology to break down regulatory clauses into data-driven, operable segments. Based on standardized templates, it automatically generates or revises regulations, ensuring that the content not only complies with regulatory requirements but also precisely adapts to specific scenarios such as electrolysis and smelting. This fundamentally addresses the challenges of poor decision-making operability and conflicting documents.

[0020] 3. Rigid closed-loop control of the execution process has been achieved, resolving the problem of inadequate supervision. Traditional supervision methods relying on manual inspections are inefficient and have many blind spots. This invention, through scenario-embedded execution, binds regulatory requirements to work points (such as key equipment and high-risk areas), and utilizes IoT sensors and AI video monitoring (such as automatic identification of violations, open flames, and intrusion into dangerous areas) for real-time comparison and early warning, transforming passive response into proactive prevention. This not only raises the regulatory compliance rate of key processes to a new high, but also, through the automatic generation of rectification orders and verification of closed loops, shifts safety management from post-event accountability to process control, greatly improving the level of inherent safety.

[0021] 4. By integrating data across the entire chain and driving intelligent decision-making, this invention solves the problem of data fragmentation. It constructs a unified data platform, breaking down data silos across regulations, systems, enforcement, and monitoring. Based on integrated multi-dimensional data, the system can automatically calculate key indicators such as compliance rate and execution rate, and can automatically identify systemic deficiencies (such as high violation rates associated with a specific clause) through data analysis, generating data-supported revision suggestions. This enables enterprises to leap from "experience-based decision-making" to "data-driven" safety management, truly meeting the core requirement of "continuous improvement" in safety production standardization. Attached Figure Description

[0022] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0023] 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. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0024] This embodiment describes an intelligent compliance management method for non-ferrous heavy metal safety standardization information. The workflow follows the principles of "data-driven, closed-loop management," sequentially executing four core stages: regulatory data acquisition and processing, system generation, scenario-based execution monitoring, and intelligent evaluation and improvement, forming a self-optimizing complete management closed loop. For example... Figure 1 As shown, the method specifically includes: constructing an industry-specific regulatory database: collecting regulatory data for the non-ferrous heavy metals industry, filtering and tagging it using AI, generating and storing tagged regulatory data; generating safety management systems: breaking down the clauses in the tagged regulatory data into structured data fragments and automatically filling them into a system template to generate safety management system documents; implementation and monitoring: associating the specific requirements in the safety management system documents with operational scenarios, and conducting real-time monitoring and closed-loop handling by comparing on-site data with system requirements to generate implementation monitoring data; evaluation and improvement: based on the tagged regulatory data and the implementation monitoring data, analyzing the compliance status and generating revision suggestions for the safety management system documents.

[0025] The method described in this embodiment is physically based on a control system. This system includes one or more servers (for deploying various software modules), network devices, and various terminal devices distributed throughout the enterprise's production site. These terminal devices include, but are not limited to: industrial computers for data acquisition, QR code identifiers and mobile smart terminals (such as explosion-proof mobile phones and tablets) for identity authentication and policy retrieval, IoT sensors for real-time monitoring (such as gas sensors and temperature sensors), smart safety helmets, and video surveillance equipment. These hardware components are connected via a network to form a complete IoT data acquisition and command execution system. The implementation process of the control method is described in detail below.

[0026] I. Building an Industry-Specific Regulatory Database: Collect regulatory data for the non-ferrous and heavy metals industry, filter and tag it using AI, and then generate and store tagged regulatory data. This is the cornerstone of achieving accurate compliance. The specific technical process is as follows.

[0027] First, multi-source data collection is performed. According to a preset task schedule (e.g., 2 AM daily), the system automatically calls publicly available data interfaces (such as RESTful APIs or WebSocket interfaces) published on the official websites of higher-level departments to obtain the latest versions, revision notes, and effective dates of relevant laws, standards, and normative documents. To improve the success rate and compliance of data collection, the crawler program incorporates rules that comply with the robots.txt protocol and uses a dynamic User-Agent to simulate browser behavior.

[0028] Then, the collected multi-source data undergoes AI-powered intelligent filtering and cleaning. AI filtering is based on a pre-defined keyword library for the non-ferrous and heavy metals industry. A semantic analysis model calculates the relevance weight between the regulatory text and the industry, filtering out general clauses with weights below a preset threshold. The collected raw text data (usually in HTML or PDF format) is first parsed and cleaned to extract the plain text content. Subsequently, the system calls a built-in semantic analysis model (e.g., a pre-trained model based on the Transformer architecture, such as BERT, fine-tuned for safety regulatory texts). This model matches the text against a pre-defined keyword library containing hundreds of non-ferrous and heavy metals industry terms (such as "splashing," "anode mud," "floating liquid," and "copper electrolysis plate"), calculating the relevance weight between the entire document or specific clauses and the industry. A configurable threshold ε (e.g., 0.07, adjustable by the administrator) is set. This threshold ε can be initially optimized based on historical data review (e.g., by statistically analyzing regulations manually marked as "relevant" and "irrelevant" over the past year to calculate the optimal cutoff point for the model score), and adaptively adjusted during system operation based on user feedback (e.g., the "relevant / irrelevant" labeling of the filtering results). General clauses with weights below this threshold ε (e.g., vague statements like "strengthen security education") are automatically filtered into a backup database, retaining only high-weight clauses in the core database. This process transforms "massive information" into "precise and useful information," reducing the average response time for enterprises to regulatory updates from 2-3 months in a manual mode to less than 2 hours in an automated system, eliminating compliance risks caused by information lag at the source.

[0029] The system automatically assigns multi-dimensional tags to the selected regulatory clauses. This is not just simple keyword matching, but a classification based on rules and a lightweight model. For example, for the clause "lead pots should have safety protection facilities" in the "Safety Production Specifications for Lead Smelting," the system will automatically assign tags such as "Metal Category: Lead, Process: Electrolytic Refining, Risk Type: High Temperature Burning, Protective Measures Type: Physical Isolation" based on its context within the "Electrolytic Refining" chapter. This tagged regulatory data is stored in JSON format, laying the foundation for subsequent accurate push notifications and associations.

[0030] II. Generating Safety Management Regulations: The clauses in the tagged regulatory data are broken down into structured data fragments and automatically populated into the regulation template to generate safety management regulation documents. The core of this step is to transform unstructured regulatory texts into structured regulations that are executable by the enterprise and internally consistent.

[0031] This invention incorporates a library of institutional templates designed based on the core elements of the "Standards for Safety Production Standardization Assessment of Non-ferrous Heavy Metal Smelting Enterprises". The templates are defined using XML or a similar structured format and include fixed blocks such as "General Provisions, Scope of Application, Division of Responsibilities, Specific Regulations (Processes, Indicators), Supervision and Assessment, and Supplementary Provisions", ensuring the standardization of the institutional format.

[0032] Clause decomposition and encoding based on Natural Language Processing (NLP). Clause decomposition is performed into structured data fragments. Specifically, NLP technology is used to identify entities, indicator values, operators, and responsible parties in the clauses, transforming unstructured text into structured data units containing clear semantic roles, thus forming the institutional gene encoding. When users need to establish a system for a new project or revision requirements, the system selects relevant clauses from tagged regulatory data and initiates the NLP process. Specifically, the system uses named entity recognition technology to identify entities in the text (such as "dust" and "protective equipment"), relation extraction technology to identify indicator values ​​(such as "0.05") and operators (such as "≤"), and semantic role labeling technology to identify responsible parties (such as "shift leader"). Ultimately, textual clauses like "dust concentration ≤ 0.05 mg / m³" are broken down into structured data fragments similar to {"Indicator Item": "Dust Concentration", "Operator": "≤", "Limit": "0.05", "Unit": "mg / m³", "Detection Frequency": "1 time / shift", "Responsible Position": "Safety and Environmental Protection Department Inspector"}. This process transforms regulatory clauses into computable and associative "institutional genetic code". These fragments adhere to a predefined JSON Schema specification, ensuring data format consistency and facilitating subsequent storage, retrieval, and calculation.

[0033] Based on the mapping relationship between clause tags and policy template blocks, the system automatically fills the aforementioned structured data fragments into the corresponding positions in the template, generating a preliminary policy draft. Subsequently, an adaptation check is initiated. An algorithm compares the text of the safety management policy document with the compliance requirements in the tagged regulatory data, automatically identifying and highlighting clause conflicts or incompatibilities across production process scenarios. The verification rule base pre-sets logical consistency rules, defined using syntax similar to the Drools rule engine (e.g., rule "Check_Approval_Authority_Conflict" when...then...end), supporting declarative business rule management. The system scans the entire policy text across processes. If it finds that the "Hot Workshop Regulations" require "approval by the safety director," while the "Hot Workshop Regulations" for the same scenario stipulate "approval by the workshop director," the system automatically identifies this conflict and highlights it. The effect is that it resolves the potential problem of "multiple departments issuing conflicting policies and inconsistent standards" before the policy is officially released, greatly improving the rigor and operability of the policy and avoiding employee confusion during implementation.

[0034] Furthermore, once a new or revised policy is approved, the system automatically initiates a distribution and training process. This process, based on the generated safety management policy document, identifies the target personnel based on its tags, pushes the document to them requiring signature, and automatically generates assessment questions. Once the target personnel pass the assessment, their relevant work permissions are activated. The system automatically determines the personnel who need to receive the policy based on policy tags (such as applicable positions and processes) and pushes an electronic version of the policy document to their mobile devices, requiring online signature. Simultaneously, based on the core clauses of the policy, the system uses NLP technology to automatically generate a set of targeted online training questions (such as multiple-choice and true / false questions) and sets a passing score. After employees complete the training and pass the assessment, the system binds the assessment results to the employee's digital identity permissions, precisely activating or disabling their specific work permissions in the background by calling the Unified Identity Authentication (IAM) interface. This process ensures that "policy revision is training," enabling seamless policy updates to be delivered to the execution level and solving the problem of policies being left unimplemented.

[0035] III. Implementation and Monitoring: The specific requirements in the aforementioned safety management regulations are linked to the operational scenarios. Real-time monitoring and closed-loop management are implemented by comparing on-site data with the regulations, generating execution monitoring data. This process ensures that regulations are seamlessly integrated into every operational step, achieving process controllability.

[0036] Work scenario binding involves associating the specific requirements in the safety management system documents with work scenarios. By setting a unique identifier at each work site, workers can scan this identifier using their mobile devices to instantly access and confirm the list of system requirements and checkpoints bound to that scenario. In the backend management interface, system administrators can drag and drop specific clauses from the generated "Safety Management System Documents" (e.g., "Forced ventilation for 30 minutes is required before confined space operations") to specific work sites on the map (e.g., "Hydrogen pipeline valve well of No. 3 smelting furnace") to bind them. The system automatically generates a unique QR code for each site. Upon arrival at the site, workers scan the QR code with their explosion-proof smartphones, and the app immediately displays the "List of System Requirements" and "Electronic Checklist" for that specific site. The beneficial effect of this approach is that it achieves precise delivery of system information from the "archive cabinet" to the "hands-on" environment, solving the problem of systems being shelved.

[0037] The system receives real-time data from on-site gas sensors (continuously monitoring CO, etc.) via an IoT platform. The system collects on-site data from various sources, including lead fume concentration, smart safety helmets (which use built-in sensors to detect whether they are worn correctly and whether people have fallen), and AI cameras (which use computer vision models to identify whether someone has entered a dangerous area and whether open flame operations are compliant). The system's built-in rule engine compares these data streams with the bound regulatory requirements in real time. For example, if the system detects that the oxygen concentration in a confined space is lower than the regulatory requirement of 19.5%, but the access control record shows that someone has entered and no high-temperature alarm has been triggered, it will immediately determine it as "violation of regulations" and generate a high-level warning.

[0038] The closed-loop handling is an automated process. Once a violation is detected, the system automatically executes the following steps: It automatically generates a rectification notice containing evidence of the violation and the relevant regulatory basis, and pushes it to the responsible person's mobile terminal; it suspends the responsible person's relevant work permissions until the responsible person uploads rectification evidence; it compares and verifies the evidence with the relevant regulatory clauses, and upon successful verification, completes the closed-loop handling and restores work permissions. Specifically: ① Generate a "Rectification Notice" containing screenshots of violations, data, and relevant clauses, and push it to the mobile terminals of the responsible team leader and area safety officer via message queue; ② Suspend the relevant work permits or equipment operation permissions; ③ The person responsible for rectification must upload photos or videos of the rectified site via the APP within the specified time, along with a written explanation of the rectification measures; ④ The system will automatically compare the uploaded rectification evidence with the relevant rules and regulations (such as image recognition to see if a safety rope is worn), or after the responsible person confirms it online, the system will mark the violation as "closed" and restore the work permission.

[0039] The aforementioned closed-loop handling process transforms safety management from post-event traceability to in-process intervention and even pre-event early warning, achieving true process control and significantly improving the implementation rate of key safety systems.

[0040] IV. Assessment and Improvement: Based on the tagged regulatory data and the execution monitoring data, analyze the compliance status and generate revision suggestions for the security management system documents. This is the core manifestation of achieving intelligent "continuous improvement".

[0041] Multi-dimensional compliance assessment. The system analyzes all data periodically (e.g., weekly) and automatically calculates key performance indicators such as "regulatory compliance rate" (the proportion of existing system provisions that cover mandatory regulatory provisions), "system implementation rate" (actual number of implementations / planned number of implementations), and "timely rectification rate of potential hazards," and displays them through a visual dashboard.

[0042] The generation of the revision suggestions specifically includes: based on the execution monitoring data, statistically analyzing the violation records and their occurrence rates related to a certain clause; when the occurrence rate of violation records related to a certain clause exceeds a preset threshold, automatically associating the corresponding requirements in the tagged regulatory data, and outputting suggestions to supplement, modify, or delete the clause.

[0043] The system continuously analyzes violation records in the "Execution Monitoring Data." Its built-in algorithm counts the number of times each policy clause is triggered and the frequency of violations. When the system detects that the violation rate related to a certain clause (such as "anti-slip measures in hydrometallurgical areas") continuously exceeds a preset threshold (e.g., 15%), it determines that the clause may be defective. Subsequently, the system automatically backtracks to the "tagged regulatory data," finds relevant regulatory basis, and combines it with a database of best practices from similar companies to intelligently generate revision suggestions (e.g., "It is recommended to change 'Caution: Anti-slip' to 'Anti-slip gratings must be laid, and 'Caution: Slippery Floor' warning signs must be set up'"), along with detailed data analysis and regulatory sources. This provides managers with data-supported decision-making basis, shortening the policy optimization cycle from months relying on experience-based judgment to weeks based on data insights, forming a self-evolving closed loop of "execution-evaluation-optimization-re-execution," meeting the core requirements of standardized assessment.

[0044] Standardized self-assessment reports can be generated with a single click. The system pre-sets a self-assessment report framework that fully corresponds to the "Standards for Safety Production Standardization Assessment of Non-ferrous Heavy Metal Smelting Enterprises." When participation in standardization assessment is required, administrators can activate the one-click generation function. The system algorithm will automatically capture and analyze historical data such as "regulatory compliance rate," "system implementation rate," and "hazard rectification rate," match them with various clauses in the assessment standards, automatically fill in evidence, calculate scores, mark deductions and reasons, and quickly generate a detailed draft "Self-Assessment Report for Safety Production Standardization Assessment." Furthermore, for identified deductions, the system can automatically associate regulatory requirements and the built-in rectification case library to generate a rectification suggestion path that includes specific rectification measures, responsible departments, and completion deadlines, greatly reducing the workload of enterprises in preparing review materials.

[0045] In summary, this invention constructs an end-to-end intelligent compliance management system through the precise collaboration of the above four stages. It deeply integrates advanced information technology with the safety management practices of the non-ferrous and heavy metals industry, achieves a high degree of automation in key processes, effectively solves many pain points of the traditional model, and provides enterprises with a solid and reliable digital security foundation.

[0046] This embodiment also provides an intelligent compliance management system for non-ferrous heavy metal safety standardization information. When the system is executed, it implements the aforementioned intelligent compliance management method for non-ferrous heavy metal safety standardization information, including: The data acquisition and processing module is used to collect regulatory data and process it through AI filtering and tagging to generate tagged regulatory data; The system generation and verification module is connected to the data acquisition and processing module. It is used to decompose the tagged regulatory data into system gene codes and fill them into the template to generate and verify the safety management system documents. The execution monitoring and closed-loop handling module is connected to the system generation and verification module, and is used to bind system requirements with work scenarios and perform real-time monitoring and closed-loop handling based on on-site data; The intelligent assessment and improvement module is connected to the data acquisition and processing module and the execution monitoring and closed-loop handling module, respectively, and is used to analyze the compliance status and generate suggestions for system revision. The policy release and training module is connected to the policy generation and verification module. It is used to push documents to target personnel, generate test questions, and manage assessments and permission activation after the policy is released.

[0047] The execution monitoring and closed-loop processing module is communicatively connected to at least one of the following: a gas sensor, a smart safety helmet, or a video monitoring device installed at the work site, to acquire on-site data in real time.

[0048] Embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may be implemented in one or more computer-usable storage media (including, but not limited to, disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0049] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0050] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0051] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0052] Contents not described in detail in this specification are prior art known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand this invention, but does not limit the scope of protection of this invention. Any equivalent substitutions, modifications, improvements, or simplifications of the above descriptions that do not depart from the essential content of this invention fall within the scope of protection of this invention.

Claims

1. A method for intelligent compliance management and control of safety standardization information for non-ferrous heavy metals, characterized in that, Includes the following steps: S1. Build an industry-specific regulatory database: Collect regulatory data for the non-ferrous heavy metals industry, and generate and store tagged regulatory data after AI screening and tagging. S2. Generate safety management system: Decompose the clauses in the tagged regulatory data into structured data fragments and automatically fill them into the system template to generate a safety management system document; S3. Execution and Monitoring: Associate the specific requirements in the safety management system documents with the work scenarios, and conduct real-time monitoring and closed-loop handling by comparing on-site data with system requirements to generate execution monitoring data; S4. Evaluation and Improvement: Based on the tagged regulatory data and the execution monitoring data, analyze the compliance status and generate revision suggestions for the security management system documents.

2. The intelligent compliance management method for non-ferrous heavy metal safety standardization information according to claim 1, characterized in that, The AI ​​screening described in S1 specifically involves: based on a pre-defined keyword library for the non-ferrous heavy metals industry, calculating the relevance weight between the regulatory text and the industry through a semantic analysis model, and filtering out general clauses with weights lower than a pre-defined threshold.

3. The intelligent compliance management method for safety standardization information of non-ferrous heavy metals according to claim 1, characterized in that, The clauses described in S2 are broken down into structured data fragments. Specifically, natural language processing technology is used to identify entities, indicator values, operators, and responsible parties in the clauses, transforming unstructured text into structured data units containing clear semantic roles, thus forming institutional gene encoding.

4. The intelligent compliance management method for safety standardization information of non-ferrous heavy metals according to claim 1, characterized in that, S2 also includes system compatibility verification, specifically: by comparing the text of the safety management system document with the compliance requirements in the tagged regulatory data through an algorithm, it automatically identifies and prompts clause conflicts or incompatible content in cross-production process scenarios.

5. The intelligent compliance management method for non-ferrous heavy metal safety standardization information according to claim 1, characterized in that, Following S2, it also includes the release and training of regulations, specifically: based on the generated safety management regulations documents, the scope of target personnel is determined according to their tags and the documents are pushed to them for signature, and assessment questions are automatically generated. After the target personnel pass the assessment, their relevant work permissions are activated.

6. The intelligent compliance management method for non-ferrous heavy metal safety standardization information according to claim 1, characterized in that, The specific requirements described in S3 are associated with the work scenario. By setting a unique identifier at the work point, the workers can scan the identifier with their mobile terminals to instantly access and confirm the list of system requirements and inspection items bound to that scenario.

7. A method for intelligent compliance management and control of non-ferrous heavy metal safety standardization information according to claim 1 or 6, characterized in that, The closed-loop processing described in S3 includes: Automatically generate a rectification notice containing evidence of violations and the relevant regulations, and push it to the responsible person's mobile terminal; The relevant work permissions of the responsible person are suspended until the responsible person uploads evidence of rectification. The evidence will be compared and verified with the relevant institutional provisions. Once the verification is successful, the closed-loop process will be completed and the operating privileges will be restored.

8. The intelligent compliance management method for non-ferrous heavy metal safety standardization information according to claim 1, characterized in that, The generation of revision suggestions described in S4 specifically includes: Based on the execution monitoring data, statistics are compiled on the violation records and their occurrence rate related to a specific clause; When the occurrence rate of violations related to a certain clause exceeds a preset threshold, the system automatically associates the corresponding requirements in the tagged regulatory data and outputs suggestions for adding, modifying, or deleting the clause.

9. An intelligent compliance management and control system for safety standardization information of non-ferrous heavy metals, characterized in that, When the system is executed, it implements the intelligent compliance management method for non-ferrous heavy metal safety standardization information as described in any one of claims 1-8, including: The data acquisition and processing module is used to collect regulatory data and process it through AI filtering and tagging to generate tagged regulatory data; The system generation and verification module is connected to the data acquisition and processing module. It is used to decompose the tagged regulatory data into system gene codes and fill them into the template to generate and verify the safety management system documents. The execution monitoring and closed-loop handling module is connected to the system generation and verification module, and is used to bind system requirements with work scenarios and perform real-time monitoring and closed-loop handling based on on-site data; The intelligent assessment and improvement module is connected to the data acquisition and processing module and the execution monitoring and closed-loop handling module, respectively, and is used to analyze the compliance status and generate suggestions for system revision. The policy release and training module is connected to the policy generation and verification module. It is used to push documents to target personnel, generate test questions, and manage assessments and permission activation after the policy is released.

10. The intelligent compliance management and control system for non-ferrous heavy metal safety standardization information according to claim 9, characterized in that, The execution monitoring and closed-loop processing module is communicatively connected to at least one of the following: a gas sensor, a smart safety helmet, or a video monitoring device installed at the work site, to acquire on-site data in real time.

Images