An AI-based intelligent management system and method for a dual prevention mechanism
The AI-based intelligent management system solves the problems of low risk identification efficiency, poor accuracy, inaccurate information delivery, and high barriers to employee participation in the existing dual prevention mechanism. It realizes the linkage between risks and hidden dangers, improves the efficiency and accuracy of risk identification, and ensures the accurate delivery and dynamic response of risk information.
Patent Information
- Application Number
- CN202611311213.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-27
- Publication Date
- 2026-11-17
Smart Images

Figure CN123117793A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of safety production risk management technology, and more specifically, to an AI-based intelligent management system and method for a dual prevention mechanism. Background Technology
[0002] The dual prevention mechanism of risk classification and control and hidden danger investigation and management (hereinafter referred to as the "dual prevention mechanism") is an important institutional arrangement in my country's work safety field. Its core lies in achieving forward-looking and source-based safety management through systematic risk identification, assessment and control, and routine hidden danger investigation and management.
[0003] However, in practice, the existing dual prevention mechanism faces the following technical bottlenecks: 1. Risk identification relies on human experience: Traditional risk identification work relies heavily on the subjective judgment of safety managers or external experts. It requires manual identification of risk points and filling out a large number of forms, which is labor-intensive, inefficient, and the consistency and accuracy of the identification results are difficult to guarantee.
[0004] 2. Disconnection between risk identification and hazard investigation: In existing systems, risk identification lists and hazard investigation and management often operate as independent modules, failing to form an effective linkage. The emergence of hazards essentially means that existing control measures are ineffective or weakened, but the system struggles to automatically identify this correlation and dynamically adjust risk levels and responsible parties.
[0005] 3. Inaccurate risk information delivery: Traditional risk lists are mostly static documents, which cannot dynamically adjust the delivery targets according to job responsibilities and risk levels. This results in "those who should know not knowing, and those who don't need to know receiving redundant information," affecting the effectiveness of control.
[0006] 4. Lagging dynamic risk response: Enterprise production activities and equipment status change in real time, but traditional assessment methods are mostly static, making it difficult to capture dynamic changes in risks and adjust control strategies in a timely manner.
[0007] 5. High barriers to employee participation: Ordinary employees lack professional knowledge in risk identification and find it difficult to actively participate in risk management work, resulting in risk identification often being limited to a few professionals and having limited coverage. Summary of the Invention
[0008] The purpose of this invention is to overcome the problems existing in the prior art and provide an AI-based intelligent management system and method for a dual prevention mechanism, so as to solve the technical problems in the prior art such as low risk identification efficiency, insufficient linkage between risks and hidden dangers, inaccurate information push, delayed dynamic response, and high threshold for employee participation.
[0009] The technical problem solved by this invention is achieved by the following technical solution.
[0010] This invention provides an AI-based intelligent management system with a dual prevention mechanism, comprising: an AI risk identification engine, a hidden danger investigation and management module, and a task supervision module. The AI risk identification engine includes: a basic information management module, a data collection module, a risk assessment calculation module, a list generation module, a measure confirmation and push module, a dynamic control level calculation module, a full-staff risk reporting module, and an AI identification and review module. The hidden danger investigation and management module includes a risk-hidden danger linkage engine and a precise push engine.
[0011] This invention also provides an AI-based intelligent management method for a dual prevention mechanism, which, when applied to the aforementioned system, includes the following steps: Step S1: Safety management personnel establish risk identification files and fill in basic information, including department, position, responsible unit, and responsible person; Step S2: Frontline staff input the equipment and facility list and work activity list for their area; Step S3: The AI risk identification engine automatically identifies the input list, identifies risk points based on the preset risk identification rule base, uses the SCL evaluation method for equipment and facilities, and uses the JHA evaluation method for work activities. Step S4: The risk assessment calculation module calculates the risk level using both the LECD method and the LSR method; Step S5: The list generation module automatically generates a suggested "Equipment and Facility Risk Classification and Control List" and "Operation Activity Risk Classification and Control List"; Step S6: The control measure confirmation and push module pushes the control measure suggestions to the application platform of the relevant personnel responsible for the risk points. The personnel responsible for the risk points check the control measure suggestions one by one and fill in the existing actual control measures. Step S7: The dynamic control level calculation module automatically calculates the control level required based on the existing actual control measures using AI intelligence; Step S8: The precise push engine pushes the risk points that need to be controlled to the corresponding responsible units and persons, and generates daily reminders on their application platforms.
[0012] The present invention has the following beneficial effects: 1. Improve the efficiency and accuracy of risk identification: Through AI intelligent automatic identification technology, the equipment list and work activity list entered by grassroots personnel are automatically converted into risk point identification results, which greatly reduces the workload of manual labor. At the same time, the consistency and accuracy of the identification results are ensured through a standardized rule base.
[0013] 2. Achieve intelligent linkage between risks and hidden dangers: Establish a dynamic linkage mechanism between risk identification and hidden danger investigation. When a hidden danger occurs and the control measures become ineffective, the system automatically raises the risk level and upgrades the responsible party for control until the hidden danger rectification is closed, thus achieving dynamic and precise risk control.
[0014] 3. Precise push and hierarchical control: Risk information is precisely pushed to regions, responsible persons, and operators, and inspection tasks of different cycles are dynamically generated according to risk changes to ensure that "those who should know are informed in a timely manner, and those who should control are effectively controlled."
[0015] 4. Lowering the barrier to employee participation: The risk reporting function for all employees allows them to supplement risk information in a simple way, which is then automatically identified and judged by AI. This not only leverages the "probe" role of frontline employees, but also ensures the quality of identification through technical means.
[0016] 5. Closed-loop management enhances execution: The system automatically generates inspection record forms and urges timely completion. Combined with the closed-loop process for hazard rectification, it forms a complete closed loop of "risk identification - measure confirmation - hazard investigation - rectification closure - risk reassessment".
[0017] 6. Dynamically adapt to the actual situation of the enterprise: The system supports the item-by-item verification and confirmation of control measures. The difference between the actual measures and the recommended measures will affect the final control level, making the risk assessment results more in line with the actual control capabilities of the enterprise. Attached Figure Description
[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a system structure block diagram according to an embodiment of the present invention.
[0020] Figure 2 This is a flowchart illustrating the risk identification and inventory generation process according to an embodiment of the present invention.
[0021] Figure 3 This is a flowchart illustrating the risk-hazard linkage control in an embodiment of the present invention.
[0022] Figure 4 This is a flowchart illustrating the process of all-employee risk reporting and review in an embodiment of the present invention. Detailed Implementation
[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Where specific conditions are not specified in the embodiments, conventional conditions or conditions recommended by the manufacturer shall apply. Reagents or instruments whose manufacturers are not specified are all conventional products that can be purchased commercially.
[0024] The following is a detailed description of an AI-based intelligent management system and method for dual prevention mechanisms provided by embodiments of the present invention.
[0025] In a first aspect, the present invention provides an AI-based intelligent management system for a dual prevention mechanism, comprising: an AI risk identification engine, a hidden danger investigation and management module, and a task supervision module. The AI risk identification engine includes: a basic information management module, a data collection module, a risk assessment calculation module, a list generation module, a measure confirmation and push module, a dynamic control level calculation module, a full-staff risk reporting module, and an AI identification and review module. The hidden danger investigation and management module includes a risk-hidden danger linkage engine and a precise push engine.
[0026] Furthermore, the AI-based intelligent management system for dual prevention mechanisms includes: The basic information management module is configured to allow safety management personnel to establish risk identification files and fill in basic information, including department, position, responsible unit, and responsible person. The data acquisition module is configured to receive risk-related data input by grassroots personnel. The risk-related data includes a list of equipment and facilities and a list of work activities within the area. The AI risk identification engine is configured to identify risk points in the input equipment and facility list and work activity list using AI intelligent automatic recognition technology, and to conduct risk assessments using SCL (Safety Checklist) and JHA (Job Hazard Analysis) evaluation methods respectively. The risk assessment calculation module is configured to perform job condition hazard analysis using the LECD method and the LSR method to calculate the risk level. The list generation module is configured to automatically generate suggested "Equipment and Facility Risk Classification and Control List" and "Operation Activity Risk Classification and Control List"; The measure confirmation and push module is configured to push the control measure recommendations to the application platform of the relevant personnel responsible for the risk points, so that they can check them one by one and fill in the existing actual control measures. The dynamic control level calculation module is configured to automatically calculate the required control level based on the existing actual control measures using AI, and push the relevant content of the risk point control to the corresponding responsible unit and person. The employee-wide risk reporting module is configured to be accessible to all employees, allowing them to receive new risk information, including information about new equipment or new work activities. The AI identification and review module is configured to automatically identify new risk points reported by employees using AI, determine whether they meet the conditions for inclusion in the risk identification database, and push eligible risk points to the relevant risk point manager for review. The hazard identification and management module is configured to include a closed-loop process for hazard identification and rectification, as well as a hazard-detection application for all employees. The risk-hazard linkage engine is configured to establish a linkage between risk identification and hazard inspection. When a hazard is discovered and before the rectification loop is closed, it is determined that there are loopholes in the existing control measures, automatically raises the control level, and pushes the control responsibility to the relevant responsible persons with higher responsibilities until the original control level is restored after the hazard rectification loop is closed. The precise push engine is configured to precisely push risk lists by region, responsible person, and operator, and generate daily, weekly, and monthly corresponding personnel inspection record forms according to the responsibility control level based on risk changes; The task supervision module is configured to urge relevant responsible persons to complete the inspection work on schedule.
[0027] Furthermore, the AI risk identification engine is equipped with a preset risk identification rule base, which is built based on historical risk data and industry standards and supports continuous optimization through machine learning algorithms.
[0028] Furthermore, the risk assessment calculation module uses the LECD method and the LSR method in parallel. The LECD method calculation formula is: R = L × E × C, where L is the probability of an accident occurring, E is the frequency of personnel exposure, and C is the severity of the consequences. The LSR method calculation formula is: R = L × S, where L is the probability of an accident occurring and S is the severity of the consequences.
[0029] Furthermore, the "Equipment and Facility Risk Classification and Control List" includes the following fields: risk point, inspection item, standard, evaluation level, control level, non-compliance with standard and consequences, control measure recommendations, control level, responsible unit, and responsible person; the control measure recommendations include engineering and technical measures, management measures, training and education measures, personal protective measures, and emergency response measures.
[0030] Furthermore, the "Work Activity Risk Classification and Control List" includes the following fields: risk point, work step, hazard source or potential event, evaluation level, control level, main consequences, control measure recommendations, control level, responsible unit, and responsible person.
[0031] Furthermore, the risk-hazard linkage engine is configured as follows: when a hazard is discovered, the risk point corresponding to the hazard and its existing control measures are identified; the correlation between the hazard type and the lack of control measures is determined; based on the degree of lack of measures, a temporary risk level is calculated according to preset rules; and the control responsibility corresponding to the temporary risk level is pushed to the responsible person at the corresponding level.
[0032] Furthermore, the system also includes a visualization module, configured to display the risk distribution of each region in the form of a four-color graph, and supports dynamic updates of risk levels.
[0033] The present invention will be further described below with reference to embodiments.
[0034] Example 1: System architecture and workflow as follows Figure 1 As shown in the figure, this embodiment provides an AI-based intelligent management system for dual prevention mechanisms, including: a basic information management module, a data collection module, an AI risk identification engine, a risk assessment calculation module, a list generation module, a measure confirmation and push module, a dynamic control level calculation module, a full-staff risk reporting module, an AI identification and review module, a hidden danger investigation and management module, a risk-hidden danger linkage engine, a precise push engine, and a task supervision module.
[0035] The system's workflow is as follows Figure 2 As shown, it includes the following steps: Step S1: Safety management personnel establish risk identification files through the basic information management module, and fill in basic information such as department, position, responsible unit, and responsible person.
[0036] Step S2: Grassroots personnel fill in the list of all equipment and facilities and the list of work activities in their area into the data collection module.
[0037] Step S3: The AI risk identification engine automatically identifies the input list and identifies risk points based on a preset risk identification rule base. For equipment and facilities, the SCL evaluation method is used; for work activities, the JHA evaluation method is used.
[0038] Step S4: The risk assessment calculation module uses the LECD method and the LSR method to calculate the risk level. The LECD method calculation formula is R = L × E × C, and the LSR method calculation formula is R = L × S. The risk level (major risk, significant risk, moderate risk, low risk) is determined based on the calculation results.
[0039] Step S5: The list generation module automatically generates suggested "Equipment and Facility Risk Classification and Control List" and "Work Activity Risk Classification and Control List". The Equipment and Facility Risk Classification and Control List includes: risk point, inspection items, standards, evaluation level, control level, non-compliance with standards and consequences, recommended control measures (engineering measures, management measures, training and education measures, personal protective equipment measures, emergency response measures), control level, responsible unit, and responsible person. The Work Activity Risk Classification and Control List includes: risk point, work steps, hazard source or potential event, evaluation level, control level, main consequences, recommended control measures (engineering measures, management measures, training and education measures, personal protective equipment measures, emergency response measures), control level, responsible unit, and responsible person.
[0040] Step S6: The control measure confirmation and push module pushes the control measure recommendations to the application platform of the relevant personnel responsible for the risk points. The responsible personnel check the control measure recommendations item by item and fill in the existing actual control measures.
[0041] Step S7: The dynamic control level calculation module automatically calculates the required control level based on existing actual control measures using AI. The degree of conformity between the actual measures and the recommended measures will affect the final control level: the more comprehensive the measures, the lower the level may be; if measures are lacking, the level may remain the same or be raised.
[0042] Step S8: The precise push engine pushes the risk points that need to be controlled to the corresponding responsible units and persons, and generates daily reminders on their application platforms.
[0043] Example 2: Risk-Hazard Linkage Control Figure 3 As shown, this embodiment details the risk-hazard linkage control process. During system operation, the hazard investigation and management module runs continuously, including a closed-loop process for hazard investigation and rectification, as well as a hazard inspection and "take a photo" application for all employees. The risk-hazard linkage engine monitors the data from the hazard investigation module in real time. When a hazard is detected, the following steps are executed: Step T1: Identify the risk points corresponding to the hidden dangers. Match the hidden dangers with the risk points using associated fields such as risk point number and location information.
[0044] Step T2: Analyze the correlation between the type of hazard and existing control measures. The formation of a hazard usually indicates a change or weakening of existing control measures, such as the failure of safety devices, non-compliance with operating procedures, or lack of personal protective equipment.
[0045] Step T3: Calculate the temporary risk level according to the degree of missing measures and preset rules. The rules can be set as follows: if a key control measure fails, the risk level will automatically increase by one or more levels; if a general measure fails, the risk level can be partially increased according to the degree of failure.
[0046] Step T4: Push the control responsibility corresponding to the temporary risk level to the responsible person at the appropriate level. For example, if a risk that was originally under workshop-level control is temporarily upgraded to company-level control due to the failure of measures caused by the hidden danger, the system will automatically push the control reminder to the company's leader in charge.
[0047] Step T5: Before the hazard is rectified and closed, the system will send daily management reminders to the person in charge of the upgraded hazard until the hazard is rectified.
[0048] Step T6: After the hazard rectification is completed and the measures are confirmed to have been repaired upon review, the system automatically restores the risk level to its original level and the control responsibility to the original responsible person. Through the above-mentioned linkage mechanism, dynamic and precise risk control is achieved, ensuring that the control level can be upgraded in a timely manner when loopholes appear in the control measures, preventing risks from getting out of control.
[0049] Example 3: Risk reporting and review by all employees Figure 4 As shown, this embodiment details the risk reporting and review process for all employees. The system provides risk reporting functionality to all employees, and the specific process is as follows: Step U1: Employees supplement new risk points through the company-wide risk reporting module. These can be newly added equipment and facilities, or newly identified work activities. Employees only need to fill in basic information, such as equipment name and work activity description.
[0050] Step U2: The AI identification and review module automatically identifies and judges the reported content. Based on a preset risk identification rule base, the system determines whether the risk point meets the entry criteria.
[0051] Step U3: For risk points that meet the entry criteria, the system pushes them to the relevant risk point manager for manual review. The manager confirms the control measure recommendations and existing control measures.
[0052] Step U4: After the review is approved, the system will incorporate the new risk points into the risk identification database and generate a control list according to the process in Example 1.
[0053] Step U5: The precise push engine pushes the newly identified risk points to all relevant personnel. The system automatically generates training tasks to ensure that relevant personnel are aware of the new risks and control requirements.
[0054] Step U6: The system records training completion status, and those who fail to complete training on time will receive a reminder. This mechanism fully leverages the "probe" role of frontline employees, while AI-assisted identification ensures the quality of risk data entry, achieving an organic combination of full employee participation and professional oversight.
[0055] Example 4: Precise Push and Inspection / Supervision This example details the precise push and inspection / supervision mechanism. The precise push engine is configured as follows: Role-based push: Based on job responsibilities and risk control levels, determine the risk information that each person should be aware of. Persons in charge and operators in the same area receive the risk list for that area; managers at different levels receive risk information corresponding to their respective control levels.
[0056] Periodic Inspection Task Generation: Daily, weekly, and monthly inspection record sheets are automatically generated according to risk level and control requirements, categorized by responsibility level. For example: Major Risk: Daily inspection, with the responsible person receiving daily reminders; Significant Risk: Weekly inspection, with the responsible person receiving weekly reminders; General Risk: Monthly inspection, with the responsible person receiving monthly reminders; Low Risk: Quarterly spot checks. Dynamic Adjustment Mechanism: When the risk level changes due to factors such as potential hazards, the inspection cycle is adjusted accordingly. During risk escalation, the inspection frequency increases accordingly; after risk downgrade, the inspection frequency returns to normal.
[0057] The task supervision module monitors and checks the completion status of tasks in real time, sends reminders to those who fail to complete tasks on time, and can also copy the information to superiors as required by management, to ensure that the inspection work is completed on schedule.
[0058] Through the above mechanism, risk management has been made routine and institutionalized, transforming the static risk list into dynamic daily inspection actions.
[0059] Example 5: System Deployment and Application This example illustrates the deployment and application of the system. The system can be deployed using a B / S architecture, supporting both cloud and local deployment modes. The front end supports access via a PC browser and also provides a mobile app or mini-program, facilitating on-site personnel to report risks, capture hazard photos, and receive inspection tasks anytime, anywhere. The system supports integration with other enterprise information systems, such as ERP systems, MES systems, personnel positioning systems, and video surveillance systems, enabling data interoperability and business collaboration. System application results show that it can significantly improve the operational efficiency of the dual prevention mechanism, reduce manual workload, improve the accuracy and timeliness of risk identification, strengthen the execution of risk control, and effectively prevent the occurrence of production safety accidents.
[0060] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. An AI-based intelligent management system with a dual prevention mechanism, characterized in that, include: The system comprises an AI risk identification engine, a hazard investigation and management module, and a task supervision module. The AI risk identification engine includes a basic information management module, a data collection module, a risk assessment calculation module, a list generation module, a measure confirmation and push module, a dynamic control level calculation module, a full-staff risk reporting module, and an AI identification and review module. The hazard investigation and management module includes a risk-hazard linkage engine and a precise push engine.
2. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, include: The basic information management module is configured to allow safety management personnel to establish risk identification files and fill in basic information, including department, position, responsible unit, and responsible person. The data acquisition module is configured to receive risk-related data input by grassroots personnel. The risk-related data includes a list of equipment and facilities and a list of work activities within the area. The AI risk identification engine is configured to identify risk points in the input equipment and facility list and work activity list using AI intelligent automatic recognition technology, and to conduct risk assessment using SCL and JHA evaluation methods respectively. The risk assessment calculation module is configured to perform job condition hazard analysis using the LECD method and the LSR method to calculate the risk level. The list generation module is configured to automatically generate suggested "Equipment and Facility Risk Classification and Control List" and "Operation Activity Risk Classification and Control List"; The measure confirmation and push module is configured to push the control measure recommendations to the application platform of the relevant personnel responsible for the risk points, so that they can check them one by one and fill in the existing actual control measures. The dynamic control level calculation module is configured to automatically calculate the required control level based on the existing actual control measures using AI, and push the relevant content of the risk point control to the corresponding responsible unit and person. The employee-wide risk reporting module is configured to be accessible to all employees, allowing them to receive new risk information, including information about new equipment or new work activities. The AI identification and review module is configured to automatically identify new risk points reported by employees using AI, determine whether they meet the conditions for inclusion in the risk identification database, and push eligible risk points to the relevant risk point manager for review. The hazard identification and management module is configured to include a closed-loop process for hazard identification and rectification, as well as a hazard-detection application for all employees. The risk-hazard linkage engine is configured to establish a linkage between risk identification and hazard inspection. When a hazard is discovered and before the rectification loop is closed, it is determined that there are loopholes in the existing control measures, automatically raises the control level, and pushes the control responsibility to the relevant responsible persons with higher responsibilities until the original control level is restored after the hazard rectification loop is closed. The precise push engine is configured to precisely push risk lists by region, responsible person, and operator, and generate daily, weekly, and monthly corresponding personnel inspection record forms according to the responsibility control level based on risk changes; The task supervision module is configured to urge relevant responsible persons to complete the inspection work on schedule.
3. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, The AI risk identification engine is equipped with a preset risk identification rule base, which is built based on historical risk data and industry standards and supports continuous optimization through machine learning algorithms.
4. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, The risk assessment calculation module uses the LECD method and the LSR method in parallel. The LECD method is calculated as follows: R = L × E × C, where L is the probability of an accident, E is the frequency of personnel exposure, and C is the severity of the consequences. The LSR method is calculated as follows: R = L × S, where L is the probability of an accident and S is the severity of the consequences.
5. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, The "Equipment and Facility Risk Classification and Control List" includes the following fields: risk point, inspection item, standard, evaluation level, control level, non-compliance with standard and consequences, control measure recommendations, control level, responsible unit, and responsible person; the control measure recommendations include engineering and technical measures, management measures, training and education measures, personal protective measures, and emergency response measures. The "Operation Activity Risk Classification and Control List" includes the following fields: risk point, operation step, hazard source or potential event, evaluation level, control level, main consequences, control measure recommendations, control level, responsible unit, and responsible person.
6. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, The risk-hazard linkage engine is configured as follows: when a hazard is discovered, it identifies the risk point corresponding to the hazard and its existing control measures; determines the correlation between the hazard type and the lack of control measures; and calculates the temporary risk level according to preset rules based on the degree of lack of measures. The control responsibilities corresponding to the temporary risk level will be pushed to the responsible persons at the corresponding level.
7. The AI-based intelligent management system for dual prevention mechanisms according to claim 1, characterized in that, It also includes a visualization module, configured to display the risk distribution of each region in a four-color graph, and supports dynamic updates of risk levels.
8. An AI-based intelligent management method with a dual prevention mechanism, applied to the system described in any one of claims 1-7, characterized in that, Includes the following steps: Step S1: Safety management personnel establish risk identification files and fill in basic information, including department, position, responsible unit, and responsible person; Step S2: Frontline staff input the equipment and facility list and work activity list for their area; Step S3: The AI risk identification engine automatically identifies the input list, identifies risk points based on the preset risk identification rule base, uses the SCL evaluation method for equipment and facilities, and uses the JHA evaluation method for work activities. Step S4: The risk assessment calculation module calculates the risk level using both the LECD method and the LSR method; Step S5: The list generation module automatically generates a suggested "Equipment and Facility Risk Classification and Control List" and "Operation Activity Risk Classification and Control List"; Step S6: The control measure confirmation and push module pushes the control measure suggestions to the application platform of the relevant personnel responsible for the risk points. The personnel responsible for the risk points check the control measure suggestions one by one and fill in the existing actual control measures. Step S7: The dynamic control level calculation module automatically calculates the control level required based on the existing actual control measures using AI intelligence; Step S8: The precise push engine pushes the risk points that need to be controlled to the corresponding responsible units and persons, and generates daily reminders on their application platforms.
9. The AI-based intelligent management method for dual prevention mechanisms according to claim 8, characterized in that, It also includes risk-hazard linkage control steps: Step T1: Identify the risk points corresponding to the hidden dangers; Step T2: Analyze the correlation between the types of hazards and existing control measures; Step T3: Calculate the temporary risk level according to the degree of missing measures and the preset rules; Step T4: Assign the control responsibilities corresponding to the temporary risk level to the responsible persons at the appropriate level; Step T5: Before the hidden danger is rectified and closed, send control reminders to the person in charge who has been upgraded to a higher level every day; Step T6: After the hazard rectification is completed and the measures are confirmed to have been repaired through a follow-up inspection, the risk level will be automatically restored to the original level, and the control responsibility will be restored to the original responsible person.
10. The AI-based intelligent management method for dual prevention mechanisms according to claim 8, characterized in that, It also includes a risk reporting and review process for all employees: Step U1: Employees supplement new risk information through the company-wide risk reporting module; Step U2: The AI identification and review module automatically identifies and judges the reported content to determine whether it meets the entry conditions; Step U3: For risk points that meet the entry criteria, push them to the relevant risk point manager for manual review and confirmation; Step U4: After approval, the new risk points will be incorporated into the risk identification database; Step U5: Push the newly identified risk points to all relevant personnel and generate training tasks; Step U6: Record the training completion status and send reminders to those who have not completed the training on time.
