A method for evaluating construction safety standards

Abstract

The application discloses a kind of building construction safety standard evaluation methods, it is related to the field of safety evaluation, comprising the following steps: step 1: extracting several construction projects to be evaluated from current construction cycle;For each construction project, define one or more safety labels for the project according to its historical data or the potential dangerous event type associated with the construction plan;Step 2: for each construction project, obtain the type, number, spatial location information of the safety emergency equipment arranged in its construction area, and the infrastructure configuration information of the area;This method converts the complex safety evaluation process into an algorithmic decision based on project labels and equipment coefficients, making the evaluation results of different projects and different periods consistent and comparable.The model can handle massive nonlinear relationships and uncover deep risk correlation patterns that are difficult for the human brain to identify directly, thereby improving the scientificity and accuracy of the evaluation.

Description

Technical Field

[0001] This invention relates to the field of safety assessment technology, specifically a method for assessing building construction safety standards. Background Technology

[0002] Construction projects are characterized by large scale, complex technology, tight schedules, and multi-disciplinary cross-operations, which objectively increases the difficulty and risk of safety management. At the same time, increasingly stringent production regulations place higher demands on enterprises' implementation of their primary responsibility for safe production. Traditional safety management models based on manual experience and periodic inspections are no longer sufficient to meet the needs of precise risk control. On the technological level, the rapid development and widespread adoption of IoT, sensors, BIM, and artificial intelligence technologies have made digital and intelligent management of construction sites possible.

[0003] Current construction safety assessments rely heavily on the personal experience of safety officers and discrete checklists, lacking quantitative standards. Conclusions from different projects or by different assessors are difficult to compare, and it's challenging to integrate objective factors such as equipment performance and environmental layout. Routine safety inspections are often periodic snapshots, failing to reflect the real-time evolution of risks arising from dynamic changes in construction. Warnings are often triggered only after hazards become apparent or even after an accident has occurred, resulting in insufficient preventative measures. Furthermore, the management information and project risk characteristic data for various safety and emergency equipment deployed on construction sites are typically separated from each other by different systems. Assessments rarely involve a systematic calculation of the comprehensive protection and response capabilities of these devices in a specific project environment, leading to unclear matching between equipment configuration and actual risks, and an inability to effectively measure input-output efficiency. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method for evaluating building construction safety standards, which can effectively solve the problems of the existing technology.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention discloses a method for assessing construction safety standards, comprising the following steps:

[0009] Step 1: Extract several construction projects to be evaluated from the current construction cycle; for each construction project, define one or more safety labels for the project based on its historical data or the types of potential hazardous events associated with the construction plan;

[0010] Step 2: For each construction project, obtain information on the type, quantity, and spatial location of the safety and emergency equipment already deployed within its construction area, as well as the infrastructure configuration information for that area;

[0011] Step 3: Based on the type, quantity, spatial location, and infrastructure configuration information of the safety and emergency equipment, evaluate the comprehensive protection coefficient and comprehensive response coefficient of the group of equipment in the current construction project environment; the protection coefficient characterizes the equipment's ability to prevent and isolate dangerous events, and the response coefficient characterizes the timeliness and coverage of the equipment's activation and operation after an event occurs.

[0012] Step 4: Construct a safety assessment model using deep learning algorithms. This model takes a specific construction project and its corresponding safety labels, the comprehensive protection coefficient and comprehensive response coefficient of the safety emergency equipment as input features, and outputs a risk coefficient that represents the overall risk level corresponding to the existing safety emergency equipment configuration scheme during the construction of the project.

[0013] Step 5: Compare the risk coefficient output by the model with the preset risk threshold. When the risk coefficient exceeds the threshold, a safety alarm will be automatically triggered for the construction project.

[0014] Furthermore, the types of safety tags in step 1 include: working at height, temporary power supply, hoisting, foundation pit excavation, scaffolding, hot work, confined space work, falling object, and mechanical injury.

[0015] Furthermore, the evaluation process for the comprehensive protection coefficient in step 3 is as follows:

[0016] Based on the factory technical parameters, product certification level, and maintenance and calibration records of safety and emergency equipment, obtain the benchmark value of its inherent protective capability against specific types of hazards;

[0017] Based on the 3D work area model of the construction project and the layout coordinates of safety and emergency equipment, the effective range of equipment is obtained, such as the spray radius of fire extinguishers, the protection area of ​​safety nets, and the field of view of monitoring cameras. The physical coverage ratio and overlap of key hazardous points, densely populated areas of workers, and dangerous paths of the project are used to generate a layout optimization score.

[0018] The impact of extreme weather on equipment stability, the obstruction of the protection range by obstacles, and the reliability of power supply and communication networks on the continuous operation of intelligent monitoring equipment are defined as adaptability correction coefficients.

[0019] The inherent protection capability benchmark value, layout optimization score, and adaptability correction coefficient are calculated using a preset weighted fusion algorithm to generate a quantitative comprehensive protection coefficient. This coefficient comprehensively reflects the static prevention capability under the combined effect of equipment performance, layout strategy, and environmental adaptability.

[0020] Furthermore, the process for generating the layout optimization score is as follows:

[0021] Based on the design drawings or on-site 3D scanning point cloud data of the construction project, a 3D spatial model is constructed that includes the work surface, equipment and facilities, personnel activity area and the location of hazard sources, and a standardized 3D model of the effective range of each type of safety emergency equipment is established.

[0022] The effective range models of each safety and emergency equipment are mapped onto the three-dimensional spatial model of the construction project based on their actual layout coordinates and orientations. Through spatial calculations, it is analyzed whether each key hazard point and high-risk operation area is fully covered by the effective range of at least one piece of equipment, and the total volume or area ratio of all key areas covered is calculated as a basic coverage index.

[0023] The system analyzes the overlap between the effective ranges of different devices, gives a positive score to the redundant coverage defined as necessary, and makes negative adjustments to the resource waste caused by overly dense device layout; at the same time, it identifies and quantifies the volume or area of ​​the protection blind zone caused by improper device layout, obstruction by obstacles, or problem of the angle of action.

[0024] Based on the aforementioned basic coverage index, overlap and redundancy assessment results, and protection blind zone quantification results, a final layout optimization score is generated through comprehensive calculation using a preset weighted scoring algorithm. This score quantifies the completeness and efficiency of the safety emergency equipment's spatial arrangement in providing preventative protection for hazardous areas of the construction project.

[0025] Furthermore, the evaluation process for the comprehensive response coefficient in step 4 is as follows:

[0026] Assess the time delay from the detection of danger signals to the full activation and readiness of relevant safety and emergency equipment, such as automatic sprinklers, alarms, emergency lighting, and isolation devices; for intelligent linkage systems, it is also necessary to assess the sequence and total time of coordinated activation of multiple devices and convert it into a timeliness score.

[0027] The simulation examines the path and time taken by on-site workers or dedicated safety officers to reach key emergency equipment operation points, such as fire hydrants, emergency stop buttons, and rescue equipment storage points, under emergency conditions. It considers the accessibility of passages, the complexity of evacuation routes, and the distribution of individual locations in the construction environment, and comprehensively derives a score for personnel accessibility and efficiency.

[0028] The timeliness score of equipment startup and system linkage, and the accessibility and efficiency score of personnel emergency operation are calculated by a preset weighted fusion algorithm to generate a comprehensive response coefficient. This coefficient comprehensively reflects the dynamic response efficiency from the occurrence of danger to the full effectiveness of protective measures.

[0029] Furthermore, the process of constructing the security assessment model in step 4 is as follows:

[0030] Collect historical construction project data to form a training dataset;

[0031] A deep learning network model framework is constructed. The input layer of the model receives the input features, performs nonlinear transformation and fusion of the features through one or more hidden layers, and finally generates a continuous risk coefficient prediction value through the output layer.

[0032] The model is trained using the training dataset. During training, the predicted risk coefficient is calculated using forward propagation, the error between the predicted value and the true target label is calculated using the loss function, and the model parameters are optimized using the backpropagation algorithm until the model converges.

[0033] The performance of the trained model is evaluated using an independent validation dataset. Based on the evaluation metrics, the model hyperparameters or network structure are adjusted to optimize the model's prediction accuracy and generalization ability, ultimately resulting in a deployable and secure evaluation model.

[0034] Furthermore, the training dataset includes: project feature vectors, corresponding safety label sets, comprehensive protection coefficients and comprehensive response coefficients of safety emergency equipment configurations as input features, and the actual safety accident level of the historical project or the manually defined risk quantification value as target labels.

[0035] Furthermore, the safety alarm notification in step 5 includes: highlighting the risky project and its location on the graphical interface of the project management platform; sending a warning message containing the risk coefficient, project identifier, and suggested inspection items to the project safety manager; and automatically activating the monitoring equipment associated with the high-risk project to conduct corresponding inspections and record video.

[0036] Furthermore, the safety assessment model trained in step 4 is automatically executed at a preset cycle or according to construction stage nodes to perform batch assessment and risk ranking of all construction projects within the cycle or stage, and generate an overall safety assessment report for that cycle or stage, which is then submitted to the preset management terminal.

[0037] (III) Beneficial Effects

[0038] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0039] 1. By introducing a safety assessment model, the complex safety assessment process that originally relied on human experience is transformed into algorithmic decision-making based on project labels and equipment coefficients. This reduces subjective arbitrariness and makes the assessment results of different projects and at different times consistent and comparable. The model can handle massive nonlinear relationships and uncover deep risk association patterns that are difficult for the human brain to recognize directly, thereby improving the scientific nature and accuracy of the assessment.

[0040] 2. By calculating risk coefficients in real time or periodically, dynamic indicators that integrate the specific hazardous attributes of the current project with the real-time performance status of on-site emergency equipment are obtained. The assessment can be continuously updated as the construction progress and equipment status change. When the risk coefficient output by the model exceeds the threshold, the alarm is triggered based on a quantitative prediction of the probability of future potential accidents, rather than a simple report of anomalies that have already occurred. This achieves early warning and makes safety intervention measures more targeted and timely. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0042] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] The present invention will be further described below with reference to embodiments.

[0045] This embodiment presents a method for assessing construction safety standards, such as... Figure 1 As shown, it includes the following steps:

[0046] Step 1: Extract several construction projects to be evaluated from the current construction cycle; for each construction project, define one or more safety labels for the project based on the types of potential hazardous events associated with its historical data or construction plan; the types of safety labels include: working at height, temporary power supply, hoisting, foundation pit excavation, scaffolding, hot work, confined space work, falling object, and machinery injury.

[0047] Step 2: For each construction project, obtain information on the type, quantity, and spatial location of the safety and emergency equipment already deployed within its construction area, as well as the infrastructure configuration information for that area.

[0048] Step 3: Based on the type, quantity, spatial location, and infrastructure configuration information of the aforementioned safety and emergency equipment, evaluate the comprehensive protection coefficient and comprehensive response coefficient of this group of equipment in the current construction project environment; the protection coefficient characterizes the equipment's ability to prevent and isolate hazardous events, and the response coefficient characterizes the timeliness and coverage of the equipment's activation and operation after an event occurs; the evaluation process for the comprehensive protection coefficient is as follows:

[0049] Based on the factory technical parameters, product certification level, and maintenance and calibration records of safety and emergency equipment, obtain the benchmark value of its inherent protective capability against specific types of hazards;

[0050] Based on the 3D work area model of the construction project and the layout coordinates of safety and emergency equipment, the effective operating range of the equipment is obtained, such as the spray radius of fire extinguishers, the protection area of ​​safety nets, and the field of view of surveillance cameras. The physical coverage ratio and overlap of key hazard points, densely populated areas of workers, and hazardous paths are then used to generate a layout optimization score. The process of generating the layout optimization score is as follows:

[0051] Based on the design drawings or on-site 3D scanning point cloud data of the construction project, a 3D spatial model is constructed that includes the work surface, equipment and facilities, personnel activity area and the location of hazard sources, and a standardized 3D model of the effective range of each type of safety emergency equipment is established.

[0052] The effective range models of each safety and emergency equipment are mapped onto the three-dimensional spatial model of the construction project based on their actual layout coordinates and orientations. Through spatial calculations, it is analyzed whether each key hazard point and high-risk operation area is fully covered by the effective range of at least one piece of equipment, and the total volume or area ratio of all key areas covered is calculated as a basic coverage index.

[0053] The system analyzes the overlap between the effective ranges of different devices, gives a positive score to the redundant coverage defined as necessary, and makes negative adjustments to the resource waste caused by overly dense device layout; at the same time, it identifies and quantifies the volume or area of ​​the protection blind zone caused by improper device layout, obstruction by obstacles, or problem of the angle of action.

[0054] Based on the aforementioned basic coverage index, overlap and redundancy assessment results, and protection blind zone quantification results, a final layout optimization score is generated through comprehensive calculation using a preset weighted scoring algorithm. This score quantitatively reflects the completeness and efficiency of the safety emergency equipment's spatial arrangement in providing preventive protection for hazardous areas of the construction project.

[0055] By introducing three-dimensional spatial modeling and quantitative calculation, the static protection effectiveness of equipment layout and the dynamic process of emergency response are transformed into coefficients that can be accurately calculated and optimized, thereby improving the scientific nature, predictability and management level of safety planning.

[0056] The impact of extreme weather on equipment stability, the obstruction of the protection range by obstacles, and the reliability of power supply and communication networks on the continuous operation of intelligent monitoring equipment are defined as adaptability correction coefficients.

[0057] The inherent protection capability benchmark value, layout optimization score, and adaptability correction coefficient are calculated using a preset weighted fusion algorithm to generate a quantitative comprehensive protection coefficient. This coefficient comprehensively reflects the static prevention capability under the combined effect of equipment performance, layout strategy, and environmental adaptability.

[0058] Step 4: Construct a safety assessment model using deep learning algorithms. This model takes a specific construction project and its corresponding safety labels, as well as the comprehensive protection coefficient and comprehensive response coefficient of the safety emergency equipment, as input features. The output is a risk coefficient representing the overall risk level corresponding to the existing safety emergency equipment configuration during the project's construction. The evaluation process for the comprehensive response coefficient is as follows:

[0059] Assess the time delay from the detection of danger signals to the full activation and readiness of relevant safety and emergency equipment, such as automatic sprinklers, alarms, emergency lighting, and isolation devices; for intelligent linkage systems, it is also necessary to assess the sequence and total time of coordinated activation of multiple devices and convert it into a timeliness score.

[0060] The simulation examines the path and time taken by on-site workers or dedicated safety officers to reach key emergency equipment operation points, such as fire hydrants, emergency stop buttons, and rescue equipment storage points, under emergency conditions. It considers the accessibility of passages, the complexity of evacuation routes, and the distribution of individual locations in the construction environment, and comprehensively derives a score for personnel accessibility and efficiency.

[0061] The timeliness score of equipment startup and system linkage, and the accessibility and efficiency score of personnel emergency operation are calculated by a preset weighted fusion algorithm to generate a comprehensive response coefficient. This coefficient comprehensively reflects the dynamic response efficiency from the occurrence of danger to the full effectiveness of protective measures.

[0062] The process of constructing the security assessment model is as follows:

[0063] Collect historical construction project data to form a training dataset; the training dataset includes: project feature vectors, corresponding safety label sets, comprehensive protection coefficients and comprehensive response coefficients of safety emergency equipment configurations as input features, and the actual safety accident level or manually defined risk quantification value of the historical project as target labels;

[0064] A deep learning network model framework is constructed. The input layer of the model receives the input features, performs nonlinear transformation and fusion of the features through one or more hidden layers, and finally generates a continuous risk coefficient prediction value through the output layer.

[0065] The model is trained using the training dataset. During training, the predicted risk coefficient is calculated using forward propagation, the error between the predicted value and the true target label is calculated using the loss function, and the model parameters are optimized using the backpropagation algorithm until the model converges.

[0066] The performance of the trained model is evaluated using an independent validation dataset. The model hyperparameters or network structure are adjusted according to the evaluation metrics to optimize the model’s prediction accuracy and generalization ability, ultimately resulting in a deployable safety assessment model. The trained safety assessment model is automatically executed at a preset cycle or according to construction stage nodes to perform batch evaluation and risk ranking of all construction projects within the cycle or stage, and generate an overall safety assessment report for the cycle or stage, which is then submitted to the preset management terminal.

[0067] By using deep learning algorithms to deeply integrate project-specific risk labels with the real-time performance coefficients of safety equipment, the real-time effectiveness of safety configurations in specific construction scenarios can be quantitatively evaluated and risk coefficients can be output, significantly improving the objectivity and accuracy of the assessment and enabling early warning of risks.

[0068] Step 5: Compare the risk coefficient output by the model with the preset risk threshold. When the risk coefficient exceeds the threshold, a safety alarm will be automatically triggered for the construction project. The safety alarm includes: highlighting the risk project and risk location on the graphical interface of the project management platform; sending a warning message containing the risk coefficient, project identifier and suggested inspection items to the project safety manager; and automatically starting the monitoring equipment associated with the high-risk project to conduct corresponding inspections and record video.

[0069] In summary, this invention upgrades traditional discrete safety inspections into continuous risk quantification assessments by integrating equipment protection coefficients and response coefficients into a safety assessment model. This improves the accuracy of the assessment, enables real-time output of risk coefficients and triggers early warnings, and provides clear decision support for on-site management personnel. It helps to allocate resources in a timely manner and eliminate hidden dangers. The model has self-optimization capabilities and can be continuously iterated as project data accumulates, ensuring that the assessment standards remain close to actual working conditions and improving the adaptability and long-term effectiveness of the method.

[0070] This invention standardizes and digitizes complex security management elements, which not only improves management efficiency but also provides a technical foundation for enterprises to establish a unified and traceable security management system, and has good scalability.

[0071] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing construction safety standards, characterized in that, Includes the following steps: Step 1: Extract several construction projects to be evaluated from the current construction cycle; for each construction project, define one or more safety labels for the project based on its historical data or the types of potential hazardous events associated with the construction plan; Step 2: For each construction project, obtain information on the type, quantity, and spatial location of the safety and emergency equipment already deployed within its construction area, as well as the infrastructure configuration information for that area; Step 3: Based on the type, quantity, spatial location, and infrastructure configuration information of the aforementioned safety and emergency equipment, evaluate the comprehensive protection coefficient and comprehensive response coefficient of this group of equipment in the current construction project environment; Step 4: Construct a safety assessment model using deep learning algorithms. This model takes a specific construction project and its corresponding safety labels, the comprehensive protection coefficient and comprehensive response coefficient of the safety emergency equipment as input features, and outputs a risk coefficient that represents the overall risk level corresponding to the existing safety emergency equipment configuration scheme during the construction of the project. Step 5: Compare the risk coefficient output by the model with the preset risk threshold. When the risk coefficient exceeds the threshold, a safety alarm will be automatically triggered for the construction project.

2. The method for assessing construction safety standards according to claim 1, characterized in that, The types of safety tags in step 1 include: working at height, temporary power supply, hoisting, foundation pit excavation, scaffolding, hot work, confined space work, falling object, and mechanical injury.

3. The method for assessing construction safety standards according to claim 1, characterized in that, The evaluation process for the comprehensive protection coefficient in step 3 is as follows: Based on the factory technical parameters, product certification level, and maintenance and calibration records of safety and emergency equipment, obtain the benchmark value of its inherent protective capability against specific types of hazards; Based on the 3D work area model of the construction project and the layout coordinates of safety and emergency equipment, the effective range of the equipment is obtained, and the physical coverage ratio and overlap of key hazardous points, densely populated areas of workers and hazardous paths of the project are used to generate a layout optimization score. The inherent protection capability benchmark value and layout optimization score are calculated using a preset weighted fusion algorithm to generate a quantitative comprehensive protection coefficient.

4. The method for assessing construction safety standards according to claim 3, characterized in that, The process of generating the layout optimization score is as follows: Based on the design drawings or on-site 3D scanning point cloud data of the construction project, a 3D spatial model is constructed that includes the work surface, equipment and facilities, personnel activity area and the location of hazard sources, and a standardized 3D model of the effective range of each type of safety emergency equipment is established. The effective range models of each safety and emergency equipment are mapped onto the three-dimensional spatial model of the construction project based on their actual layout coordinates and orientations. Through spatial calculations, it is analyzed whether each key hazard point and high-risk operation area is fully covered by the effective range of at least one piece of equipment, and the total volume or area ratio of all key areas covered is calculated as a basic coverage index. Analyze the overlap between the effective ranges of different devices, give a positive score to the redundant coverage defined as necessary, and make a negative adjustment for the resource waste caused by overly dense device deployment; Based on the basic coverage index, overlap and redundancy assessment results, and protection blind zone quantification results, a final layout optimization score is generated through comprehensive calculation using a preset weighted scoring algorithm.

5. The method for assessing building construction safety standards according to claim 1, characterized in that, The evaluation process for the comprehensive response coefficient in step 4 is as follows: Assess the time delay from the perception of a hazard signal to the full activation and readiness of relevant safety and emergency equipment, and convert it into a timeliness score; Simulate the path and time for on-site workers or dedicated safety officers to reach the key emergency equipment operation points in an emergency. Consider the accessibility of passages, the complexity of evacuation routes, and the distribution of individual locations in the construction environment, and comprehensively derive a score for personnel accessibility and efficiency. The timeliness score of equipment startup and system linkage, and the accessibility and efficiency score of personnel emergency operation are calculated using a preset weighted fusion algorithm to generate a comprehensive response coefficient.

6. The method for assessing construction safety standards according to claim 1, characterized in that, The process of constructing the security assessment model in step 4 is as follows: Collect historical construction project data to form a training dataset; A deep learning network model framework is constructed. The input layer of the model receives the input features, performs nonlinear transformation and fusion of the features through one or more hidden layers, and finally generates a continuous risk coefficient prediction value through the output layer. The model is trained using the training dataset. The performance of the trained model is evaluated using an independent validation dataset, ultimately resulting in a deployable security evaluation model.

7. The method for assessing construction safety standards according to claim 6, characterized in that, The training dataset includes: project feature vectors, corresponding safety label sets, comprehensive protection coefficients and comprehensive response coefficients of safety emergency equipment configurations as input features, and the actual safety accident level of the historical project or the manually defined risk quantification value as target labels.

8. The method for assessing construction safety standards according to claim 1, characterized in that, The safety alarm notification in step 5 includes: highlighting the risky project and its location on the graphical interface of the project management platform; sending a warning message containing the risk coefficient, project identifier and suggested inspection items to the project safety manager; and automatically activating the monitoring equipment associated with the high-risk project to conduct corresponding inspections and record video.

9. The method for assessing construction safety standards according to claim 1, characterized in that, The safety assessment model trained in step 4 is automatically executed at a preset cycle or according to construction stage nodes. It performs batch assessment and risk ranking of all construction projects within the cycle or stage, and generates an overall safety assessment report for that cycle or stage, which is then submitted to the preset management terminal.

Images